genetic studies on recurrent miscarriage
TRANSCRIPT
GENETIC STUDIES ON RECURRENT MISCARRIAGE
Milja Kaare
Folkhälsan Institute of Genetics
and
Department of Medical Genetics
University of Helsinki,
Helsinki, Finland
ACADEMIC DISSERTATION
To be publicly discussed, with the permission of
the Medical Faculty of the University of Helsinki,
in Biomedicum Helsinki, Haartmaninkatu 8, lecture hall 2,
on February 13th
2009, at 1 pm.
Helsinki 2009
Supervised by
Docent Kristiina Aittomäki, M.D., Ph.D.
Department of Clinical Genetics,
Helsinki University Central Hospital
Folkhälsan Institute of Genetics and
Department of Medical Genetics,
University of Helsinki
Reviewed by
Docent Leena Anttila, M.D., Ph.D.
Väestöliitto Fertility Clinics Ltd, Turku Clinic
Department of Obstetrics and Gynecology,
University of Turku
Docent Nina Horelli-Kuitunen, Ph.D., medical geneticist
Medix Laboratories Ltd, Espoo
Official opponent
Professor Seppo Heinonen, M.D., Ph.D.
Department of Obstetrics and Gynecology
University of Kuopio and
Kuopio University Hospital
ISBN 978-952-92-5027-1 (paperback)
ISBN 978-952-10-5259-0 (PDF)
http://ethesis.helsinki.fi
Yliopistopaino
Helsinki 2009
Contents
List of original publications _____________________________________________ 6
Abbreviations ________________________________________________________ 7
Abstract _____________________________________________________________ 9
Introduction ________________________________________________________ 11
Review of the literature ________________________________________________ 12
1 Early human embryonic development _______________________________ 12
1.1 Fertilization __________________________________________________ 12
1.2 Implantation __________________________________________________ 12
1.3 Placentation __________________________________________________ 14
1.4 Gastrulation and organogenesis ___________________________________ 14
2 Recurrent miscarriage ____________________________________________ 15
2.1 Identifiable causes of miscarriage _________________________________ 16
2.1.1 Risk factors for miscarriage __________________________________ 19
2.2 Genetic factors causing miscarriage _______________________________ 19
2.2.1 Chromosomal abnormalities __________________________________ 20
2.2.2 Skewed X chromosome inactivation ___________________________ 21
2.2.3 Y chromosome microdeletions ________________________________ 21
2.3.4 Gene mutations and polymorphisms ___________________________ 22
2.2.5 Aberrant placental gene expression ____________________________ 24
2.2.6 Epigenetic mechanisms _____________________________________ 25
3 Identification of genetic causes underlying RM ________________________ 26
3.1 Identification of disease genes ____________________________________ 27
3.1.1 The Human Genome Project _________________________________ 27
3.1.2 Traditional identification of disease genes _______________________ 28
3.1.3 The candidate gene approach _________________________________ 28
3.1.4 Mutation screening _________________________________________ 29
3.1.5 Confirming the nature of a new genetic variation _________________ 30
3.2 Mouse as a model organism _____________________________________ 30
3.3 Candidate genes for RM ________________________________________ 32
3.3.1 Amnionless _______________________________________________ 32
3.3.2 Thrombomodulin and Endothelial Protein C Receptor _____________ 35
3.3.3 p53 _____________________________________________________ 38
3.3.4 The mitochondrial genome ___________________________________ 39
Aims _______________________________________________________________ 43
Materials and methods ________________________________________________ 44
1 Study subjects ___________________________________________________ 44
1.1 Patients (I, II, III, IV, V) ______________________________________ 44
1.2 Controls (I, II, III, IV, V) ______________________________________ 44
1.3 Placental samples (III, IV) _____________________________________ 45
1.4 Ethical issues (I, II, III, IV, V) __________________________________ 45
2 Methods ________________________________________________________ 45
2.1 Genealogic analysis (I) _______________________________________ 45
2.2 DNA extraction (I, III) ________________________________________ 45
2.3 Candidate gene analysis _______________________________________ 46
2.4 Immunohistochemical analysis of p53 (III) ________________________ 48
2.5 X chromosome inactivation analysis (V) __________________________ 48
2.6 Y chromosome microdeletion analysis (V) ________________________ 48
2.7 Statistical analysis (I, II, III, IV, V) ______________________________ 49
Results and discussion ________________________________________________ 50
1 Genealogic studies (I) _____________________________________________ 50
2 Candidate gene analysis (I, II, III, IV) _______________________________ 52
2.1 Amnionless (I) ________________________________________________ 52
2.1.1 Variations in the Amnionless gene _____________________________ 52
2.2 Thrombomodulin and Endothelial Protein C Receptor (II, unpublished) ___ 54
2.2.1 Variations in the TM gene ___________________________________ 54
2.2.2 Variations in the EPCR gene _________________________________ 55
2.2.3 TM/EPCR variations and outcome of heparin treatment ____________ 56
2.3 p53 (III, unpublished) __________________________________________ 57
2.3.1 Variations in the p53 gene ___________________________________ 57
2.3.2 p53 expression in RM placenta ________________________________ 58
2.4 Mitochondrial genes (IV) ________________________________________ 59
2.4.1 Variations in the mitochondrial genome _________________________ 60
2.4.2 Analysis of placental samples _________________________________ 62
3 The impact of polymorphisms on phenotype __________________________ 63
3.1 Variations in the nuclear genome (I, II, III, unpublished) _______________ 63
3.1.1 Possible splice site variations _________________________________ 65
3.1.2 Couple analysis _____________________________________________ 66
3.2 Variations in the mitochondrial genome (IV) ________________________ 67
4 Analysis of sex chromosome characteristics (V) _______________________ 68
4.1 X chromosome inactivation analysis _______________________________ 68
4.2 Y chromosome microdeletion analysis _____________________________ 71
5 Limitations of the study ___________________________________________ 73
Conclusions _________________________________________________________ 74
Acknowledgements ___________________________________________________ 77
References __________________________________________________________ 79
6
List of original publications
This thesis is based on the following publications, which are referred to in the text by their
Roman numerals. In addition, some unpublished results are presented.
I Kaare M, Painter JN, Ulander VM, Kaaja R, Aittomaki K (2006) Variations
of the Amnionless gene in recurrent spontaneous abortions. Mol Hum
Reprod. 12:25-9.
II Kaare M*, Ulander VM*, Painter JN, Ahvenainen T, Kaaja R, Aittomaki K
(2007) Variations in the Thrombomodulin and Endothelial Protein C
Receptor genes in couples with recurrent spontaneous abortions. Hum
Reprod. 22:864-8.
III Kaare M, Bützow R, Ulander V-M, Kaaja R, Aittomäki K, Painter JN
(2009). Study of p53 gene mutations and placental expression in recurrent
miscarriage cases. RBM Online. E-pub ahead of print on 9 January.
IV Kaare M, Götz A, Ulander V-M, Ariansen S, Kaaja R, Suomalainen A,
Aittomäki K. Do mitochondrial mutations cause recurrent miscarriage?
Submitted.
V Kaare M, Painter JN, Ulander V-M, Kaaja R, Aittomäki K (2008) Sex
chromosome characteristics and recurrent miscarriage. Fertil Steril. 90:2328-
33.
* These two authors contributed equally to this work.
Publication II also appears in the thesis of Veli-Matti Ulander (2007)
The articles are reprinted with the permission of their copyright holders.
7
Abbreviations
ACE angiotensin I-converting enzyme
AMN amnionless
APC activated protein C
APP amyloid beta precursor protein
AR androgen receptor
ATP adenosine triphosphate
AZF azoospermia factor
BAD BCL2 antagonist of cell death
BAX BCL2-associated X protein BID BH3-interacting domain death agonist
bp base pair
cDNA complementary DNA
CI confidence interval
CO2 Cytochrome c oxidase subunit II
cpm count per minute
CS carnegie stage
CTLA-4 cytotoxic T lymphocyte-associated 4
CUBN cubilin
CytB cytochrome b
dHPLC denaturing high performance liquid chromatography
DNA deoxyribonucleic acid
E embryonic day
EAA European Academy of Andrology
EMQN European Molecular Genetics Quality Network
EPCR endothelial protein C receptor
ESE exonic splicing enhancer
FasL Fas ligand
FSH-RO follicle stimulating hormone-resistant ovaries
HGP human genome project
HLA human leukocyte antigen
ICM inner cell mass
IFN-γ interferon-γ
IGS Imerslund-Gräsbäck syndrome
IL6/10 interleukin 6/10
ITGB3 integrin beta-3
MDM2 mouse double minute 2
MELAS mitochondrial encephalopathy with lactic acidosis and stroke-
like episodes
MERRF myoclonic epilepsy and ragged-red fibres
MLPA multiplex ligation-dependent probe amplification
MLS microphthalmia with linear skin defects syndrome
8
mRNA messenger RNA
mtDNA mitochondrial DNA
MTHFR methylenetetrahydrofolate reductase
MUC1 mucin 1
ND2 NADH dehydrogenase subunit 2
OFD1 oral-facial-digital syndrome type 1
OR odds ratio
PAI-1 plasminogen activator inhibitor 1
PCR polymerase chain reaction
PGC primordial germ cells
PGD preimplantation genetic diagnosis
Q-PCR quantitative real-time PCR
PP14 placental protein 14
RM recurrent miscarriage
RNA ribonucleic acid
rRNA ribosomal RNA
SIFT sorting intolerant from tolerant
SNP single nucleotide polymorphism
SR-protein Ser/Arg-rich-protein
STS sequence tagged sites
Th-1 type 1 T-helper cell
TM thrombomodulin
tRNA transfer RNA
UTR untranslated region
VEGF vascular endothelial growth factor
VTE venous thromboembolism
XCI X chromosome inactivation
9
Abstract
Recurrent miscarriage (RM) is defined as three or more consecutive pregnancy failures
and is estimated to affect ~1% of couples trying to conceive. The cause of RM remains
unknown in approximately 50% of cases although much work has been done in the past
to identify the underlying mechanisms. In this study, it was hypothesized that some of the
underlying factors yet to be discovered are genetic.
The aim of this study was to obtain new information on genetic causes of RM. The aim
was to search for mutations in genes known to cause miscarriage in animal models and
thereby find new genetic causes for unexplained miscarriages in humans. Four candidate
genes – Amnionless (AMN), Endothelial protein C receptor (EPCR), Thrombomodulin
(TM), and p53 – were chosen for the study as they have all been considered to be crucial
for normal embryonic development in the mouse. In addition, the mitochondrial genome
was studied because mitochondria are involved in processes important in early
development. Furthermore, sex chromosome characteristics suggested to underlie
miscarriage were also studied.
A total of 40 couples and 8 women with unexplained RM were collected for this study.
The patients were screened for mutations in the AMN, TM and EPCR genes using dHPLC.
Six interesting exonic and potential splice site disrupting variations were detected in these
genes, namely c.363G>A, c.843+11C>T, c.1339_1344dup, and c.1170-6C>T in AMN,
c.1728+23_+40del in TM, and c.323-9_336dup in EPCR. Two of these variations,
c.363G>A and c.843+11C>T were more frequent in RM women compared to controls.
However, their phenotypic effects cannot be determined without further investigations.
While most variations in the AMN, TM and EPCR genes were identified in both patients
and controls and are likely to be silent polymorphisms, others may have functional
significance or modifier effects in combination with mutations or polymorphisms in other
genes.
An association between the C11992A polymorphism of the p53 gene and RM was
detected. The results indicate that women carrying the C/A or A/A genotype have a
twofold higher risk for RM than women with a C/C genotype. This strengthens the results
of previous studies reporting that p53 sequence variations may cause miscarriage. The role
10
of variation C11992A in embryonic development is, however, difficult to predict without
further studies. The role of p53 in RM was also investigated through expression analysis
of p53 in placental tissue. The results showed normal expression of p53 in the samples
studied.
When screening the mitochondrial genome a heteroplasmic mtDNA variation was
found in 13 (27%) RM women and 9 (19%) control women, which was unexpected as
heteroplasmic variations are reported to be rare. One novel variation within the coding
regions of the mitochondrial genome was detected. In addition, 18 previously reported
polymorphisms were identified. Although the detected variations are likely to be neutral
polymorphisms, a role in the aetiology of miscarriage cannot be excluded as some mtDNA
variations may be pathogenic only when a threshold is reached.
Recent publications have reported skewed X chromosome inactivation and Y
chromosome microdeletions to be associated with RM. Therefore, these sex chromosome
abnormalities in the context of RM were investigated. No association between skewed X
chromosome inactivation or Y chromosome microdeletions and RM in the Finnish
patients studied were detected.
In addition, data on ancestral birthplaces of the patients were collected to study any
possible geographic clustering, which would indicate a common predisposing factor. The
results showed clustering of the birthplaces in eastern Finland in a subset of patients. This
suggests a possibility of an enriched susceptibility gene which may contribute to RM.
These results encourage further studies to determine new genetic factors contributing to
the aetiology of RM.
11
Introduction
Miscarriages are the most common complication of pregnancy, affecting approximately
15% of all clinically recognized pregnancies in the general population. The exact
frequency of miscarriages is, however, unknown as miscarriages frequently occur before
the woman is aware of her pregnancy. It is estimated that more pregnancies are lost
spontaneously than are actually carried to term (Rai and Regan, 2006; Stephenson and
Kutteh, 2007). Most of the miscarriages are sporadic and non-recurrent, and are often
caused by chromosome abnormalities in the foetus (Dhont, 2003). Recurrent miscarriage
(RM), defined as three consecutive pregnancy failures, is estimated to affect ~1% of all
couples trying to conceive (Li et al., 2002; Rai and Regan, 2006).
There are numerous factors that may cause RM, but the underlying problem often
remains undetected. Although much work has been done to identify the underlying
mechanisms, the cause of miscarriage can be identified in only ~50% of cases. The known
causes of RM include chromosomal and metabolic abnormalities, uterine anomalies, and
immunologic factors (Li et al., 2002; Rai and Regan, 2006). Even though RM is a
heterogeneous condition and the progress in identifying causative factors has been slow,
the repetitive pregnancy losses in some couples and the high percentage of unexplained
RM indicate that there are specific underlying causes yet to be found.
This study was conducted to gain further knowledge concerning genetic causes of RM.
Identification of the underlying factors is crucial for the development of more successful
treatment and improvement of the outcome of future pregnancies in women experiencing
RM. Genetic factors causing RM are, however, difficult to study because the foetus is lost
at an early stage of development and is therefore difficult to examine. Consequently, most
of the studies conducted on RM in general, and in this thesis work, are based on studying
the couples experiencing the miscarriages. We hypothesized that several genetic factors
such as mutations in genes, either identified from animal models or regulating functions
important in different stages of early foetal development, or abnormalities in the sex
chromosomes may underlie the miscarriages in the couples studied.
12
Review of the literature
1 Early human embryonic development
1.1 Fertilization
The development of a human being starts from a single cell, a zygote, which results from
an ovum being fertilized by a spermatozoon. Fertilization occurs in the fallopian tube
within 48 hours of ovulation. After fertilization, a series of symmetrical cell divisions
create a mass of 12 to 16 totipotent cells, the morula, which is still enclosed within the
zona pellucida (Norwitz et al., 2001; Sariola et al., 2003). The morula enters the uterine
cavity approximately three days after fertilization. The appearance of a fluid filled inner
cavity within the mass of cells marks the transition from a morula to a blastocyst and is
accompanied by cellular differentiation (Norwitz et al., 2001; Wang and Dey, 2006). The
first differentiation event gives rise to trophoblasts, specialized epithelial cells that give
rise to extraembryonic structures, including the placenta. The remaining cells segregate at
one pole of the embryo to form the inner cell mass (ICM) (Figure 1). Within three days of
entering the uterine cavity, the embryo hatches from the zona pellucida, thereby exposing
its trophoblast cells, which enables implantation (Norwitz et al., 2001; Sariola et al., 2003;
Wang and Dey, 2006).
1.2 Implantation
The newly formed zygote has a discrete time frame in which it must prepare itself for
implantation. Preparation begins upon oocyte fertilization and takes approximately 6 to 7
days. During this time successive cell division occurs, and the outer trophoblast cells of
the blastocyst differentiate into cytotrophoblasts which diverge into villous and invasive
subtypes forming the building blocks of the placenta (Norwitz et al., 2001; Vitiello and
Patrizio, 2007). At the same time, when the blastocyst prepares itself for implantation, the
uterus becomes receptive. Uterine receptivity is defined as the state during the period of
endometrial maturation and proliferation when the blastocyst can become implanted.
Multiple signals synchronize the development of the blastocyst and the preparation of the
uterus. Of the many aspects of the synchronization process, the role of steroid hormones
is
13
the best understood. Peptide hormones, growth factors, and cytokines also have roles in
the process. The cascade of signalling events that occur in both foetal and maternal tissues
establishes an appropriate environment critical to the development and survival of the
foetus (Norwitz et al., 2001; Wang and Dey, 2006; Vitiello and Patrizio, 2007).
Implantation occurs in three stages. The first stage is the initial adhesion of the
blastocyst to the uterine wall. The next stage is characterized
by increased physical
interaction between the blastocyst and the uterine epithelium. Shortly thereafter invasion
begins and syncytiotrophoblasts, the outer layer of the trophoblasts, penetrate the uterine
epithelium (Wang and Dey, 2006; Chavatte-Palmer and Guillomot, 2007; Vitiello and
Patrizio, 2007). By the 10th day after conception, the blastocyst is completely embedded
in the stromal tissue of the uterus as the uterine epithelium has regrown to cover the site of
implantation (Norwitz et al., 2001). The interaction between an activated blastocyst and a
receptive uterus is part of a complex process that leads to implantation
and the early stages
of placental development. Failure
to synchronize the processes involved in these
interactions results in failure of implantation (Cross et al., 1994; Wang and Dey, 2006;
Chavatte-Palmer and Guillomot, 2007).
Figure 1. The early stages of human development. A) An unfertilized egg cell is fertilized in the fallopian
tube by a sperm cell leading to B) a fertilized cell containing both the maternal and paternal pronuclei. The
cell undergoes cleavage to C) the 2-cell stage, D) the 4-cell stage and E) the 8-cell stage, which is reached
two to three days after fertilization. F) The morula enters the uterine cavity approximately three days after
fertilization. G) Implantation of the blastocyst, consisting of a blastocyst cavity, trophoblast cells and the
inner cell mass, into the uterine epithelium is initiated around seven days after fertilization. Modified from
Wang and Dey, 2006.
14
1.3 Placentation
The placenta is the first organ to form during mammalian embryogenesis. The placenta is
a vital organ without which the embryo cannot survive in the uterine environment. The
function of the placenta during early gestation is primarily to mediate implantation of the
embryo into the uterus. After implantation, the major function of the placenta is to
mediate, as well as to regulate, nutrient uptake from the mother to the foetus. It also
establishes the interface for gas exchange between the maternal and foetal circulation.
Another important function of the placenta is the regulation of the maternal immune
response so that the foetal semi-allograft is tolerated during pregnancy. The placenta also
acts as an important source of hormones and growth factors that are needed for the initial
maternal recognition of pregnancy (Cross et al., 1994; Cross, 2006; Sadler et al., 2006).
Because the placenta is critical for survival it is very sensitive to disruption. Genetic or
environmental factors that affect the development of the placenta are associated with poor
pregnancy outcome. Abnormal expression of specific regulatory genes in the placenta and
abnormal functions of imprinted genes may cause placental dysfunction (Coan et al.,
2005). Abnormalities in placental functions can lead to a variety of problems including
implantation failure, placental insufficiency, foetal growth retardation and embryonic
death. Complications that become apparent relatively late in pregnancy may actually
reflect errors that occurred already during placental development (Cross et al., 1994;
Rossant and Cross, 2001; Cross, 2006).
1.4 Gastrulation and organogenesis
While the syncytiotrophoblasts start to penetrate into the wall of the uterus during
implantation, the embryoblast/ICM also continues developing. The embryoblast forms a
bilaminar embryo, composed of the epiblast and the hypoblast. The hypoblast forms the
yolk sac. The epiblast undergoes gastrulation at approximately day 16 following
fertilization. Gastrulation establishes the three germ layers—the endoderm, ectoderm, and
mesoderm—all of which will give rise to various organ systems. The mesoderm also
interacts with the trophoblast tissue to form the umbilical cord (Cross et al., 1994; Sadler
et al., 2006). At this point the actual embryonic phase starts and lasts from the third week
until the eighth week following fertilization. The organ systems differentiate at greatly
varying rates during this phase. For example, the circulatory system is largely functional at
15
the end of this period, whereas the nervous system continues to undergo massive cell
division and is only beginning to establish functional connections (Sadler et al., 2006).
The remainder of human development, from weeks nine to thirty-eight, is called the foetal
period. During this period the embryo acquires its human appearance. The foetal period is
characterized by rapid growth and continued tissue and organ differentiation (Sariola et
al., 2003; Sadler et al., 2006).
2 Recurrent miscarriage
Human reproduction entails a fundamental paradox: although it is critical for the survival
of the species, the process is relatively inefficient. Maximal fecundity (the probability of
fertilization during one menstrual cycle) is approximately 30%. In addition, although much
research and many advances have been made in reproductive medicine during the past
decades, miscarriage, the spontaneous loss of a pregnancy before the foetus reaches
viability, remains the most common complication of pregnancy (Clark et al., 2001; Pandey
et al., 2004).
There are two types of miscarriage; sporadic and recurrent. In the general population
~15% of all clinically recognized pregnancies end in miscarriage (Stray-Pederson and
Stray-Perderson, 1984; Nybo Andersen et al., 2000). However, the rate of miscarriage
may be as high as 50-60% if the undetected, very early pregnancy losses occurring within
the first weeks following gestation are taken into account (Roberts and Lowe, 1975;
Wilcox et al., 1988). Of the pregnancies that are lost, ~70-75% are caused by failures of
implantation or early placentation and are therefore not clinically
recognized as
pregnancies. As a consequence, only about half of all conceptions advance into the second
trimester of pregnancy (Figure 2) (Rai and Regan, 2006).
RM is estimated to occur in 1-2% of all couples (Tulppala et al., 1993; Katz and
Kuller, 1994), but the exact prevalence of RM is dependent on its definition. RM is often
defined as the occurrence of three or more consecutive, clinically detectable pregnancy
failures before the 20th
week of gestation (Li et al., 2002b; Quenby et al., 2002; Pandey et
al., 2004). However, some studies have also included patients with two miscarriages
(Laskin et al., 1997; Stephenson et al., 1998) and in some studies the miscarriages were
16
not necessarily consecutive. This increases the prevalence of RM up to 5% of all couples
trying to conceive (Rai and Regan, 2006).
Figure 2. Number of conceptions expected to reach different stages of pregnancy. Of all fertilized oocytes
only about half result in a livebirth. Most of the pregnancies are lost during early stages of development.
Modified from Rai and Regan, 2006.
2.1 Identifiable causes of miscarriage
The repetitive miscarriages in some couples suggest that certain women are at particular
risk of losing their pregnancy and that there must be an underlying explanation for this.
Even though much work has been done to identify these underlying mechanisms, the
aetiology of miscarriage is still unknown in many cases. Additionally, on an individual
level the exact reason for a particular miscarriage is rarely defined. Despite the extensive
medical testing and experimental treatments that many patients undergo, the cause often
remains unclear (Plouffe et al., 1992; Clifford et al., 1994; Carrington et al., 2005). The
main identifiable causes of miscarriage include the following categories (Table 1):
Uterine pathology; Uterine defects, such as uterine anomalies and fibroids, can
predispose to miscarriage by affecting implantation (Bulletti et al., 1996; Bajekal and Li,
2000).
Endocrine abnormalities; Following implantation, the maintenance of the pregnancy is
dependent on a range of endocrinological events that will eventually aid in the successful
growth and development of the endometrium and the foetus (Potdar and Konje, 2005;
Arredondo and Noble, 2006). Normal endometrial development and maturation is
dependent on steroid hormones, particularly oestrogen and progesterone. Therefore,
Conseption
Implantation
Recogniced
pregnancy
Livebirth
100
80
60
40
20
%
17
miscarriage can be a result of subnormal hormone production or abnormal endometrial
response to circulating steroid hormones (Li et al., 2002a; Arredondo and Noble, 2006).
In addition, several other maternal endocrinological abnormalities such as uncontrolled
diabetes (Mills et al., 1988), high androgen levels (Okon et al., 1998; Bussen et al., 1999),
hyperprolactinaemia (Tal et al., 1991), thyroid dysfunction (Roberts and Murphy, 2000),
and obesity (Wang JX et al., 2002) have been implicated as aetiological factors for RM.
Immunological factors; Since the conceptus displays paternal gene products and
antigens, it is possible that the maternal immune system recognizes these as foreign,
resulting in an immune response (Dalton et al., 1998; Hill and Choi, 2000). Survival of the
semiallogenic foetus is dependent on suppression of the maternal immune response and
miscarriage may be a consequence of a failure of this suppression. It is assumed that both
maternal and embryonic regulating factors protect the embryo against an adverse maternal
immunological reaction. Maternal adaptation to the implanting embryo is needed for
successful establishment of the foetal-placental unit, and the embryo must be able to
stimulate an immunological reaction from the mother in order to become protected from
Table 1. Causes of RM and their estimated frequencies in women with RM (Stephenson and Kutteh, 2007;
Warren and Silver, 2008).
Factor Frequency
Uterine defects ~15%
Anatomic deformation
Uterine fibroids
Cervical incompetence
Endocrinologic ~10%
Abnormal hormone production
Diabetes
Thyroid dysfunction
Obesity
Immunologic ~15%
Aniphospholipid syndrome
Maternal autoimmune factors (acquired or inherited)
Thrombophilic ~10%
Prothrombotic state (acquired or inherited)
Environmental ~5%
Infections
Teratogens (alcohol, drugs, tobacco)
Parental chromosome ~5%
abnormality
Balanced translocations
18
cytotoxic molecules (Bulletti et al., 1996; Li et al., 2002b; Pandey et al., 2004). The
mechanism by which the foetus escapes rejection by the maternal immune system is
unknown. However, differences in the concentration of circulating immunocompetent
cells, such as natural killer (NK) cells, and cytokines have been reported between RM
women and controls (Wegmann et al., 1993; Reinhard et al., 1998; Ntrivalas et al., 2001;
Yamada et al., 2001).
Prothrombic state; A prothrombotic state can be broadly defined as any condition,
whether inherited or acquired, that predisposes to venous or arterial thrombosis (Hiatt and
Lentz et al., 2002). It is postulated that maternal thrombophilia is a risk factor for
miscarriage because a successful pregnancy is dependent on satisfactory placental
development and sustained placental function is based on placental circulation (Kujovich,
2004). These processes require the establishment of an adequate foeto-maternal
circulatory system, which in turn may be disturbed by a prothrombotic state. Factor V
Leiden, mutations in the prothrombin gene, and protein C and protein S deficiencies are
hereditary prothrombotic disorders of clinical importance and associations between
increased risk of miscarriage and these thrombophilic disorders have been reported
(Preston et al., 1996; Rey et al., 2003; Kujovich, 2004).
One of the major acquired prothrombotic disorders is antiphospholipid antibody
syndrome, in which the body recognizes its own phospholipids as foreign and produces
antibodies against them. Lupus anticoagulant and anticardiolipin antibody are the two
known antiphospholipid antibodies that are associated with RM (Levine et al., 2002). It is
unknown exactly how the antiphospholipid antibody syndrome adversely affects
pregnancy, but one theory is that it may cause blood clots in the blood vessels of the
placenta, impairing placental function (Rand et al., 1997). However, antiphospholipid
syndrome is a treatable cause of RM. Various treatments, such as heparin, aspirin, and
intravenous immunoglobulin, have been used in attempts to improve the pregnancy
outcome of women with antiphospholipid syndrome (Empson et al., 2002; Farquharson et
al., 2002). Heparin is a thromboprophylactic agent that can bind to antiphospholipid
antibodies, thereby protecting the trophoblast and maternal vascular endothelium from
damage in early pregnancy. Later in pregnancy, when the intervillous circulation has been
established, heparin helps to decrease the risk of placental thrombosis and infarction
(Girardi, 2005, Bose et al., 2005).
19
Environmental factors; Heavy metals, organic solvents, and ionizing radiation are
confirmed teratogens, and exposure to these can contribute to pregnancy loss (Dhont,
2003). Alcohol and cocaine are also confirmed teratogens, while caffeine and smoking are
suspected teratogens but their teratogenic impact is still controversial (Dominguez-Rojas
et al., 1994; Parazzini et al., 1998; Cnattingius et al., 2000; Kesmodel et al., 2002).
2.1.1 Risk factors for miscarriage
In addition to identifiable causes of miscarriage, other factors increasing the risk for
miscarriage have been reported. Increased maternal age is reported to be an independent
risk factor for miscarriage (Risch et al., 1988; Abdalla et al., 1993; Nybo Andersen et al.,
2000). The risk of mscarriage increases rapidly after the age of 35. The risk of a
miscarriage is 9% in women aged 20-24 years, while in women aged over 45 years the
risk is as high as 75% (Nybo Andersen et al., 2000). This is explained partly, but not
totally, by the association of maternal age with increased likelihood of chromosomal
abnormalities in the foetus (Munne et al., 1995; Snijders et al., 1999).
The outcome of previous pregnancies, especially of the first pregnancy, is also an
independent predictor of future pregnancy outcome. For young women who have never
experienced pregnancy loss the rate of miscarriage is as low as 5% (Regan et al., 1989).
The risk increases to approximately 30% for women with three or more losses but with a
previous live-born infant, and up to 50% for women with no live-born infants (Poland et
al., 1977), indicating that the risk of miscarriage increases with the number of previous
miscarriages. Additionally, a normal karyotype in a pregnancy loss is a predictor of
subsequent miscarriages because an abnormal foetal karyotype is associated with sporadic
foetal loss. The frequency of abnormal foetal karyotypes significantly decreases with the
number of miscarriages (Ogasawara et al., 2000).
2.2 Genetic factors causing miscarriage
In addition to the previously described identifiable causes of miscarriage, genetic factors
are a major cause of clinically recognized miscarriages. In the following sections, the
currently reported genetic causes associated with miscarriages are described.
20
2.2.1 Chromosomal abnormalities
Most unbalanced chromosome aberrations result in severe phenotypes already in early
pregnancy and therefore lead to miscarriage. Consequently the incidence of foetal
unbalanced chromosomal abnormalities gradually decreases with duration of pregnancy to
less than 1% among live-born children (Stern et al., 1996; Stephenson et al., 2002).
Overall, 50-70% of all sporadic miscarriages are associated with cytogenetic abnormalities
in the foetus. The most frequent of these abnormalities is trisomy, followed by polyploidy,
monosomy X, and unbalanced translocation (Kalousek et al., 1993; Stephenson et al.,
2002). The most common trisomy is trisomy of chromosome 16, followed by trisomy of
chromosomes 21 and 22 (Stephenson and Kutteh, 2007). These chromosomal
abnormalities may be inherited or may arise de novo in the germ cells due to aberrant
oocyte spindle formation and meiotic division (Devine et al., 2000; Kuo, 2002).
In women with RM the rate of chromosomal abnormalities in the foetus is lower than
in women with sporadic miscarriage. The miscarriages in women with RM may be due to
de novo numerical chromosome abnormalities, in particular autosomal trisomies of
chromosomes 13, 14, 15, 16, 21, and 22, and monosomy X (Pandey et al., 2005; Warren
and Silver, 2008). In women with RM, however, the miscarriage may also be due to
recurrent chromosomal anomalies in the foetus resulting from a balanced aberration in one
of the parents inherited by the offspring in an unbalanced form. Cytogenetic screening of
couples with RM has revealed that parental chromosomal abnormalities occur in either
partner in 5-7% of couples with RM, while the rate in the normal population is
approximately 0.2% (Fryns and Buggenhout, 1998). Balanced translocations in either
parent, including reciprocal and Robertsonian translocations, are the most common
abnormalities, detected in about 4% of couples with RM. The other chromosomal
abnormalities include inversions and sex chromosome abnormalities (Tulppala et al.,
1993; Clifford et al., 1994; Fryns and Buggenhout, 1998). Even though the theoretical risk
for abnormal offspring for phenotypically normal carriers of balanced translocations is
approximately 50%, preimplantation genetic diagnosis (PGD) analysis has shown that the
risk varies greatly. For carriers of a balanced translocation the risk of conceiving a
chromosomally abnormal embryo, due to abnormal segregation at meiosis, varies from 20-
80%. The reproductive risk conferred by chromosome rearrangements is dependent on the
size, location, and type of rearrangement and whether it is carried by the woman or her
male partner (Munne et al., 2000; Otani et al., 2006; Lim C et al., 2008).
21
2.2.2 Skewed X chromosome inactivation
X chromosome inactivation (XCI) is defined as the process where one of the two X
chromosomes in each somatic cell of a healthy female is inactivated during early
embryogenesis. The process is important to achieve appropriate dosage compensation for
X chromosomal genes (Lyon, 1961; Brown and Robinson, 2000). Females are therefore
mosaics for two cell types; cells with the paternal X chromosome as the active
X and cells
with the maternal X chromosome as the active X. Normally the process occurs randomly
on both the maternal and paternal X chromosome, resulting in a 50:50 distribution of the
two cell types. A significant deviation from this distribution is called skewed XCI (Puck
and Willard, 1998; Minks et al., 2008). Recently, several studies have shown that women
experiencing RM exhibit nonrandom XCI more often than control women (Lanasa et al.,
1999; Uehara et al., 2001; Kuo et al., 2008). Kuo et al. (2008) reported extremely skewed
XCI in 7.7% of RM women and in 1.3% of control women (p=0.01).
Skewed XCI, defined as preferential inactivation of one of the two X chromosomes,
may be the result of a chance event, or due to genetic factors involved in the X inactivation
process (Brown and Robinson, 2000). In some cases a non-random inactivation pattern
may be an indicator of an abnormality in the inactivated X, e.g. X chromosome
aberrations such as microdeletions or mutations in X chromosomal genes (Belmont, 1996;
Uehara et al., 2001). Deviation from random inactivation rates can also lead to the
expression of recessive X-linked traits in heterozygous females, who are expected to be
healthy carriers of the disease gene (Azofeifa et al., 1995). It has been proposed that X-
linked mutations which are lethal in males cause skewed XCI in female carriers. For
example, microphthalmia with linear skin defects syndrome (MLS) and oral-facial-digital
syndrome type 1 (OFD1) are previously described X-linked dominant male-lethal diseases
(Franco and Ballabio, 2006). Such mutations may result in loss of male conceptuses and
increased frequency of miscarriage in carrier females (Lanasa et al., 1999; Uehara et al.,
2001).
2.2.3 Y chromosome microdeletions
The majority of the testing for RM assesses the woman. The male partners, however, have
been less thoroughly investigated. Semen analysis is not generally included in the initial
assessment for RM. Yet, sperm integrity is required for sperm–egg interactions,
22
fertilization, and early embryonic development and sperm quality is associated with the
embryo’s ability to reach the blastocyst stage and progress to implantation (Simerly et al.,
1995; Van Blerkom, 1996; Moomjy et al., 1999).
Microdeletions of the long arm of the human Y chromosome are associated with
variable spermatogenic failure (Vogt et al., 1996; Hopps et al., 2003). These
microdeletions occur in regions called AZFa, AZFb, and AZFc (more recently named
AZFa, P5/Proximal-P1, P5/Distal-P1, P4/Distal-P1, b2/b4-AZFc), containing a number of
genes (Hopps et al., 2003; Ferlin et al., 2007). Although ten recurrent deletions in the AZF
regions have been described in detail (Ferlin et al., 2007), surprisingly little is known
about the function of the individual genes and transcription units located within these
regions (Hopps et al., 2003; Foresta et al., 2005). Y chromosome microdeletions are found
in approximately 7% of men with very low sperm counts (Ferlin et al., 2007). These men
also often show a higher incidence of sperm chromosomal abnormalities (Foresta et al.,
2005). An elevated rate of sperm chromosome abnormalities has also been found in some
men of RM couples (Rubio et al., 1999; Egozcue et al., 2000). A recent study reports that
the incidence of sperm chromosome abnormalities is 16.5% in male partners of women
with RM, while the incidence is under 5% in controls (Al-Hassan et al., 2005).
Interestingly, a pilot study conducted to determine if Y chromosome microdeletions are
associated with RM, reported that a significant portion (82%) of male partners of women
with RM had Y chromosomal microdeletions, which may underlie the miscarriages
experienced by their partners (Dewan et al., 2006).
2.3.4 Gene mutations and polymorphisms
Mutations and polymorphisms in several genes are suggested to contribute to RM. Many
of these sequence variations are located in genes involved in immunological mechanisms
or blood coagulation pathways, but also genes involved in other functions have been
associated with RM.
Since the conceptus displays paternal gene products and antigens, it is possible that the
maternal immune system recognizes these as foreign, resulting in an immune response
(Dalton et al., 1998; Hill and Choi, 2000). In recent years, several genes involved in the
immune response have been associated with pregnancy loss. The CTL-A gene is a
23
regulator of T-cell activation and an A/G polymorphism in exon 1 of the gene has been
reported to be associated with RM in the Chinese population. The frequency of the GG
genotype in RM patients was significantly higher compared with controls (48.8% versus
33.3%, p = 0.011) (Wang et al., 2005). Polymorphisms in the major histocompatibility
complex have also been associated with RM. A polymorphism in the HLA-G promoter
region, the HLA-G *0104 and *0105N alleles (Aldrich et al., 2001; Ober et al., 2003) and
the HLA-DRB1 *1505 allele have been associated with the immunopathogenesis of RM
(Takakuwa et al., 2003). In addition, a number of cytokine gene polymorphisms are
associated with pregnancy loss. Cytokines are critical immunoregulatory molecules which
play a crucial role in early embryo development and placentation. Associations have been
shown between IL-6, IL-10, and IFN-γ gene polymorphisms and RM (Daher et al., 2003;
Costeas et al., 2004; Prigoshin et al., 2004; von Linsingen et al., 2005).
It is assumed that maternal thrombophilia is a risk factor for miscarriage because a
successful pregnancy is dependent on an adequate foeto-maternal circulatory system,
satisfactory placental development and sustained placental function. These functions may
be disturbed by a prothrombotic state. Factor V and prothrombin gene mutations, protein
C and protein S deficiencies, cause hereditary prothrombotic disorders (Preston et al.,
1996; Kujovich, 2004). Associations between these thrombophilic disorders and an
increased risk of miscarriage have been reported. Depending on the study the Factor V
Leiden mutation and the prothrombin G20210A mutation are reported to increase the risk
for RM between 2-fold and 5.5-fold, and between 2.5-fold and 4.9-fold, respectively
(Souza et al., 1999; Foka et al., 2000; Reznikoff-Etiévan et al., 2001; Rey et al., 2003).
Protein C and protein S deficiencies are much rarer and they have been reported to
increase the risk of early RM ~3.5-fold and ~14.5-fold, respectively (Rey et al., 2003;
Kujovich, 2004). The MTHFR gene has an important role in homocysteine metabolism
and missense mutations C677T and A1298C are known to cause hyperhomocysteinemia.
Hyperhomocysteinemia is a risk factor for vascular thrombophilia and has also been
shown to associate with RM (Couto et al., 2005; Mtiraoui et al., 2006). In addition,
polymorphisms in the ITGB3 gene, which encodes a human platelet antigen, may increase
the risk of miscarriage by causing thrombophilia. The PLA2 polymorphism in the ITGB3
gene was shown to increase the risk of foetal loss 3.6-fold (Ruzzi et al., 2005).
In addition, various other genes have been associated with pregnancy loss. For
example, mutations in the SCO2 gene, needed for cytochrome c oxidase assembly, were
24
detected in a compound heterozygous state in aborted foetuses and are suggested to cause
miscarriage (Tay et al., 2004). A novel A/G polymorphism in intron 6 of the eNOS gene,
which is active during implantation, was recently reported to be associated with RM
(Suryanarayana et al., 2006). A polymorphism in the VEGF gene, a well characterized
regulator of angiogenesis, was also recently reported as a new susceptibility factor for
RM. Homozygosity of the VEGF –1154A variation was more frequent in RM women
compared to controls and increased the risk of RM 2.7-fold (Coulam and Jeyendran,
2008).
2.2.5 Aberrant placental gene expression
The growth and differentiation of placental structures are regulated by many genes and the
placenta is very sensitive to genetic disruption, as reflected by the increasing list of
targeted mouse mutations in genes such as Mash2 and Hand1 that cause placental defects
and embryonic lethality (Rossant and Cross, 2001). Genetic abnormalities that affect the
development of the placenta can cause placental insufficiency, foetal growth retardation
and embryonic death, both in mice and humans (Rossant and Cross, 2001; Cross, 2006).
It is assumed that normal placentation and normal placental functions are mediated by
differential
gene expression in the placenta. It has also been suggested that genes
expressed in the chorionic villi act as critical regulators of placentation and hence of early
human development (Dizon-Townson et al., 2000; Baek, 2004). Recently, over 30 genes
that are expressed aberrantly in pregnancy failure have been identified. The genes showing
different levels of expression between controls and patients with RM can be mainly
grouped as immunity related, angiogenesis related, and apoptosis related genes (Baek et
al., 2002; Choi et al., 2003). Higher expression levels of apoptosis related genes such as
caspase 3, 6, 7, 8, 9, 10, 12, BAD, BAX, BID, Fas and FasL were shown
in chorionic villi
from patients with RM compared to controls (Choi et al, 2003). Immunity related genes
with aberrant gene expression in chorionic villi include PP14, MUC1 (Baek et al., 2002),
CD95 (Hoshimoto et al., 2002), and Annexin II (Aarli et al., 1997). These genes may be
involved in immunological foeto-maternal reactions. Fibronectin, Integrin and VEGF are
examples of angiogenesis related genes showing aberrant gene expression in RM placenta
(Choi et al, 2003). These genes are involved in the formation of blood vessels that are
crucial for placental and embryonic development.
25
2.2.6 Epigenetic mechanisms
The term epigenetics refers to processes that regulate gene activity without affecting
the
DNA sequence and are heritable through cell division. Epigenetic modifications of DNA
such as methylation are important for genome function during development, and
disturbances in the highly coordinated epigenetic processes contribute to developmental
failures (Lucifero et al., 2004; Santos and Dean, 2004). In most mammalian species,
including humans, there are at least two developmental periods during which methylation
patterns are reprogrammed genome wide; germ cell development and preimplantation.
After genome wide demethylation of the genome in developing germ cells, remethylation,
which resets the parent-of-origin specific marks, occurs. The second phase of
reprogramming of methylation occurs during the preimplantation stage. The paternal
genome is actively demethylated within a few hours of fertilization in the human zygote,
whereas the maternal genome is passively demethylated by a replication dependent
mechanism after the two cell embryo stage. After the demethylation stage, de novo
methylation of the genome occurs in the developing embryo. This reprogramming is likely
to have a crucial role in establishing nuclear totipotency required for normal development
(Reik et al., 2001; Reik and Dean, 2001; Haaf, 2006).
Genomic imprinting is a mechanism which results in differential expression of
maternal and paternal genes. Only one of the inherited gene copies is active in an embryo
depending on the parental origin of the gene. DNA methylation in germ cells is critical for
imprinting as most imprinted genes have differentially methylated regions (Reik and
Dean, 2001; Swales and Spears, 2005). Anomalous expression of imprinted genes during
development is sometimes caused by imbalanced representation of maternal and paternal
genetic contributions, uniparental disomy, and is associated with developmental delay or
gestational loss (Mutter, 1997).
It is estimated that there are several hundred imprinted genes in the human genome. It
has been argued that imprinted genes play essential roles in controlling
the placental
supply of maternal nutrients to the foetus, by regulating the growth of the placenta and/or
the activity of
transplacental transport systems. In some cases,
errors in genomic
imprinting are embryo lethal while in others they lead to developmental disorders and
diseases (Reik and Dean, 2001; Lucifero et al., 2004; Swales and Spears, 2005).
26
3 Identification of genetic causes underlying RM
Despite numerous investigation the cause of miscarriage remains unknown in
approximately 50% of cases (Plouffe et al., 1992; Tulppala et., al 1993, Clifford et al.,
1994). A number of possible aetiologies have been proposed to explain the RM in these
cases. The hypotheses, however, change over time and new findings have enriched our
understanding of possible mechanisms underlying RM. For example, specific HLA-G
alleles (Aldrich et al., 2001) and skewed X-inactivation (Lanasa et al., 1999; Uehara et al.,
2001) are newly factors identified to be associated with RM and could, in some cases, be
the cause for unexplained RM. However, additional genetic factors contributing to the
aetiology of RM are likely to be detected.
There are different mechanisms by which genetic factors may contribute to
miscarriage, and different modes of inheritance by which these mutations and
polymorphisms are transmitted from parents to foetus. Dominant de novo mutations
arising in the germ cells of the parents may result in affected foetuses and cause
miscarriage. Alternatively, de novo mutations during an early developmental stage in the
foetus may result in mosaicism, which also may cause a severe phenotype resulting in
miscarriage (McFadden and Friedman, 1997; Los et al., 2004). De novo mutations are
difficult to study because these mutations cannot be detected by studying DNA samples
from parents and foetal/placental samples are rarely available.
Autosomal recessive mutations cause a number of genetic diseases, and are common
especially in isolated populations (Peltonen et al., 1999; Norio, 2003a; Teebi and Teebi,
2005; Weinstein, 2007). The role of recessive mutations in specific genes in RM can be
studied without foetal samples by screening the parents for heterozygous mutations, which
when inherited by the foetus in a homozygous state are expected to be lethal. In isolated
populations disease causing mutations are often enriched and may also show uneven
geographical distribution. Evidence of such geographical clustering can be studied using
genealogic data (Kere, 2001; Norio, 2003a).
Not only autosomal mutations, but also abnormalities in the sex chromosomes can
underlie RM. X-linked mutations are predicted to be capable of inducing foetal loss. Some
X-linked recessive mutations are expected to be lethal to all male foetuses inheriting the
mutation and thereby result in miscarriage (Lanasa et al., 1999). Recently, a study
27
suggested an association between Y-chromosome microdeletions and RM (Dewan et al.,
2006). The role of these genetic factors in the aetiology of RM can be further studied
using parental DNA samples.
In addition to the nuclear genome, the mitochondria contain their own DNA, which is
transmitted only from the mother to the foetus. The human mitochondrial genome is
16,569 bp in size and encodes 37 genes (DiMauro, 2001). Mutations in these genes
leading to dysfunctional proteins may have a role in RM as the mitochondria are involved
in processes necessary for early development (Van Blerkom, 2004).
It is also possible that mutations in imprinted genes can cause miscarriage. When the
imprinting mechanism is malfunctioning loss of function or inappropriate expression can
occur in intact genes and may result in miscarriage (Reik and Dean, 2001; Swales and
Spears, 2005). The role of imprinted genes in RM is, however, difficult to study as
imprinting still is a poorly understood phenomenon.
Recently, several genes such as CTL-A, IL6, ITGB3, and VEGF have been shown to be
associated with RM (Wang et al., 2005; von Linsingen et al., 2005; Ruzzi et al., 2005;
Coulam and Jeyendran, 2008) and it is likely that other genes, in which mutations and
polymorphisms cause or increase the risk for miscarriage, will be identified in the near
future. Therefore, further studies and screening of candidate genes are required to establish
the role of genetic factors in RM.
3.1 Identification of disease genes
3.1.1 The Human Genome Project
The human genome consists of approximately 3 billion base pairs (bp) and is estimated to
contain 20 000-25 000 genes (Balakrishnan et al., 2005; Antonarakis and Beckmann,
2006). The Human Genome Project (HGP) began in 1990 and its main goals were to
determine the nucleotide sequence of the human genome and to identify all human genes.
A draft sequence of the human genome was published in 2001 (Lander et al., 2001; Venter
et al., 2001). An updated sequence, published in 2003, covered ~99% of the euchromatic
genome (International Human Genome Sequencing Consortium 2004). The first complete
28
genome sequence of a single human individual was published in 2007 (Levy et al. 2007).
In addition to the human genome, the genoms of hundreds of other organisms have been
sequenced (www.ebi.ac.uk/genomes/index.html). This genomic information has
dramatically changed the process of identifying disease genes. It provides new, invaluable
tools for understanding the basic human genetic make-up and how variations in the
genomic sequence result in disease (Peltonen and McKusick, 2001;
http://genomics.energy.gov).
3.1.2 Traditional identification of disease genes
Before the complete human genome sequence was available, disease genes were
traditionally identified using functional and positional cloning methods. In functional
cloning the gene is isolated based on information regarding the protein or its function. The
method can be used if the amino acid sequence of a protein is at least partially known
(Collins, 1992). However, most often, the protein encoded by the disease gene is not
known and the positional cloning method is used to identify a gene based on the
knowledge about its physical location in the genome. Using polymorphic markers and
DNA samples from families with affected members the disease gene locus is localized to a
genomic region by linkage analysis. Disease gene identification based on positional
cloning has been widely used in the research of monogenic diseases (Collins, 1992;
Collins, 1995).
3.1.3 The candidate gene approach
The candidate gene approach is an important step towards disease gene identification and
relies on information about known genes. New effective methods for mutation screening
together with the information offered by the HGP have made the candidate gene approach
a commonly used method to search for disease genes as information on physical locations
and sequences of many genes is available. The candidate gene approach can be used when
the biochemical background of the defect of interest is known, when a chromosomal
region has been linked to a disease or if an animal model of a disease has been established
(Collins, 1992; Ballabio, 1993; Zhu and Zhao, 2007). The mouse is a widely used model
organism for studying mammalian gene function and genes causing a phenotype in the
mouse can be used as candidate genes in human disease studies (Strachan and Read,
29
2004). Interaction partners of identified proteins defective in a disease can also be
considered as candidate genes for the disease of interest. In addition, candidate genes can
be selected based on results from genome wide expression profiling data showing
differences in gene expression between cells or tissues from affected and control
individuals (Antonarakis and Beckmann, 2006).
3.1.4 Mutation screening
When a candidate gene is selected, the next step in disease gene identification is searching
for mutations in the gene under study. During the past years several different methods
have been used for mutation screening. To date, one of the most commonly used method
is direct sequencing. In addition, denaturing high-performance liquid chromatography
(dHPLC) (Xiao and Oefner 2001) and multiplex ligation-dependent probe amplification
(MLPA) (den Dunnen and White, 2006) are frequently used to screen DNA samples for
sequence variations due to their sensitivity and specificity. A set of patients and control
DNA samples are screened for mutations segregating with the disease phenotype. It is
important to screen a set of healthy control individuals to be able to assess the significance
of the nucleotide changes and to confirm that the variation truly is associated with the
disease (Strachan and Read, 2004).
When screening candidate genes different types of sequence variations can be
identified. Nonsense mutations are point mutations that introduce premature stop codons
into the coding regions. Nonsense mutations usually lead to the degradation of the
transcript or may produce truncated nonfunctional protein products. Missense mutations
are point mutations that result in the substitution of an amino acid and are sometimes
difficult to differentiate from rare neutral polymorphisms. A missense mutation is more
likely to be pathogenic if it changes an amino acid that is conserved. If the change alters
the chemical nature of the side chain of the amino acid this usually suggests a pathogenic
role for the variation. In addition, sequence variations in parts of the gene that code for
functionally important domains are more likely to be pathogenic. Deletions, insertions,
and duplications can disrupt the reading frame, resulting in a completely different protein
translation. In addition to frameshift mutations, deletions, insertions, and duplications
can introduce premature stop codons or alter the protein by amino acid(s) deletion or
insertion (Strachan and Read, 2004).
30
Splicing mutations affect the consensus sequences at splice donor, splice acceptor, or
splice branch sites and may abolish the splicing partially or completely. In addition,
cryptic splice sites can be activated or splicing enhancers and silencers altered. These
types of splicing mutations may lead to complete or partial exon skipping or intron
retention (Cartegni et al. 2003; Baralle and Baralle, 2005). In addition, neutral and silent
mutations/polymorphisms are likely to be observed in candidate gene screening. A
neutral mutation is a variation that results in the substitution of a different, but
chemically similar amino acid. Silent mutations are variations that do not result in a
change to the amino acid sequence. They may occur either within an exon or within the
non-coding regions of the gene (Strachan and Read, 2004). Polymorphisms in the non-
coding regions of the genome, including the promoter region, the 5´- and 3´-untranslated
regions, and the intragenic regions are also commonly detected. Some of these variations
may affect gene splicing, and transcription and translation, and thereby the expression of
the gene (Wang et al., 2006).
3.1.5 Confirming the nature of a new genetic variation
Identifying a disease gene is a possible starting point to study the function of the protein
encoded by the gene. A candidate disease gene has to be carefully studied to determine if
there is enough evidence that the identified changes are not just neutral polymorphisms
but do cause the disease under study. The candidate gene can be demonstrated to be the
disease gene in different ways. Mutation screening is usually the first step in the process.
Identification of mutations in several unrelated individuals strongly suggests that the
candidate gene is causing the disease, but formal proof requires additional evidence. To
prove that the candidate gene has a pathogenic role in the disease functional studies, such
as restoration of the normal phenotype in vitro, or production of an animal model for the
disease, are required (Strachan and Read, 2004).
3.2 Mouse as a model organism
Genetic approaches in model organisms provide a powerful means by which to examine
the biological basis of human diseases. Over the past century, the mouse has become the
most used animal model system for genetic research because of its close genetic and
31
physiological similarities to humans. In addition, its small size and short generation time
have allowed large scale mutagenesis experiments and extensive genetic crosses. Many
mouse mutations have been established as reliable models of human disease while others
have provided insight into complex developmental and physiological pathways in
mammals. Targeted mutagenesis approaches in the mouse have led to dramatic increases
in our understanding of human disease processes (Brown, 1998; Brown and Nolan, 1998;
Nolan et al., 2002). There are a number of mouse models described in the literature and
databases on different mutant strains are available on the internet
(http://ceolas.org/VL/mo/; Peters et al., 2007).
Sequencing of the mouse genome was initiated in 1999 and the first draft sequence
was published in 2002 by Waterston and coworkers. Because of the comparatively high
level of sequence conservation between human and mouse coding sequences, almost all
human genes have an easily identifiable mouse homolog. Therefore, disease genes
identified in mouse models are usually good candidate genes for human diseases. The
mouse and human genomes both appear to contain 20 000-25 000 protein coding genes
and more than 90% of the mouse genome can be aligned with a region of the human
genome due to conserved synteny (Waterston et al., 2002). In addition to providing
important data on mouse genes and the structure of the mouse genome, the sequence data
have also been valuable in comparative genomics. One application of human-mouse
comparative sequence analysis involves the identification of orthologous genes.
Functionally important genomic regions are commonly homologous in human and mouse
genomes and genetic information obtained from the mouse can thereby be used for human
disease studies (Collins et al., 1998; Nobrega and Pennacchio, 2004).
Apart from the very early stages of preimplantation development, human embryos are
inaccessible for research and studies are limited to diseased aborted foetuses. Due to the
difficulties of studying implantation and foetal development in humans, much of our
knowledge of the genes involved in these processes has been derived from animal models.
There are several animal models used to gain information about human development
including; nematode worm, fruit fly, rat, guinea pig, rabbit, pig, sheep, cow, and non-
human primates (Lee and DeMayo, 2004; Dvash et al., 2006; Carter, 2007). The most
commonly used model organism is the mouse, but there are, however, significant
differences in the mode of implantation, placental structure, yolk sac, number and function
of placental hormones, and particularly in the length of pregnancy between mouse and
32
human (Rossant and Cross, 2001; Carter, 2007). Despite these differences, gene ablation
in the mouse has proven to be a powerful tool in elucidating gene function during
embryonic development (Rossant and Cross, 2001; Dvash et al., 2006).
Knock-out mouse models have been used to identify a number of genes required for
normal implantation and placentation (Rossant and Cross, 2001;Lee and DeMayo, 2004;
Carter, 2007). In addition, imprinted genes required for normal embryo development have
been studied in mouse models (Mutter, 1997; Lucifero et al., 2004). A number of mouse
models, in which deletions of a specific gene are embryo-lethal at a specific
developmental stage, have also been described (Healy et al., 1995; Sah et al., 1995; Wang
et al., 1996; Gu et al., 2002). These mouse models can be used to identify candidate genes
for human RM. A gene identified in mouse to be embryo lethal can be considered as a
clinically important candidate gene for human RM if the mouse and human genes have a
highly homologous sequence and similar normal functions in both species.
3.3 Candidate genes for RM
The following sections describe four candidate genes for human RM. They were selected
based on results from studies on animal models. In addition, the genes encoded by the
mitochondrial genome were considered as candidate genes because the mitochondria are
involved in processes important for development but their role in early development is still
unknown.
3.3.1 Amnionless
The amnionless mouse model
The amnionless mouse has a recessive prenatally lethal insertional mutation in the Amn
gene (Wang et al., 1996). The gene was named Amnionless (Amn) as the first visibly
striking feature of the mutant mouse was the lack of an amnion surrounding the foetus.
Examination of the mutant embryos at different times of gestation indicated that the
mutation causes a gastrulation stage (embryonic day (E) 6.5-E8.0) defect in the
homozygous mutant mouse (Wang et al., 1996; Tomihara-Newberger et al., 1998).
Although gastrulation begins normally in the Amn mutants they remain small and
undifferentiated and generate no amnion. The growth and differentiation of the embryonic
33
ectoderm cells are severely impaired in the mutant embryos. The first signs of abnormal
development in embryos homozygous for the transgene insertion can be seen between
E6.5 and E7.0 and the mutant embryos die and are resorbed between E9.5 and E10.5
(Table 2) (Wang et al., 1996).
Functions of the Amn gene
The Amn gene encodes a type I transmembrane protein (Kalantry et al., 2001). During the
early
post-implantation stages mouse Amn is expressed exclusively in the visceral
endoderm (Tomihara-Newberger et al., 1998; Kalantry et al., 2001). In adults, both mouse
and human Amn is expressed mainly in the kidney and the intestinal epithelium (Tanner et
al., 2003; Strope et al., 2004).
The exact role of the Amn protein in mouse development has not yet been determined,
but the gene function is required in the visceral endoderm for normal primitive streak
formation (Tomihara-Newberger et al., 1998). The visceral endoderm is an epithelial cell
layer that does not contribute directly to the foetus, but is needed to absorb
and digest
nutrients from the maternal environment, to provide signals required for the organization
of the body axes, and to secrete factors that promote correct positioning of the
primitive
streak (Beddington and Robertson, 1999; Martinez-Barbera and Beddington, 2001; Perea-
Gomez et al., 2002). At the onset of gastrulation the middle region of the primitive streak
is absent in the Amn mutant mouse (Tomihara-Newberger et al., 1998).
In wild-type mouse embryos Amn co-localizes with two other gene products, cubilin
(Cubn) and megalin, at the surface of the visceral endoderm. In the Amn mutant mouse
Cubn fails to localize to the surface of the visceral endoderm. This indicates that Amn is
an essential component of the Cubn receptor complex, and it has been suggested that the
Amn/Cubn-complex is required for endocytosis/transcytosis of one or more ligands
in the
visceral endoderm. (Strope et al., 2004). In humans, AMN and CUBN also form a
complex. This complex is required for uptake of vitamin B12 (Fyfe et al., 2004; Luder et
al., 2008). Mutations in either of these genes cause selective intestinal malabsorption of
vitamin B12, a rare autosomal recessive disorder known as Imerslund-Gräsbeck syndrome
(IGS) or megaloblastic anemia 1 (Imerslund, 1960; Gräsbeck et al,. 1960; Aminoff et al.,
1999; Tanner et al., 2004).
34
The fact that the Amn protein has a similar structure in both mice and humans suggests
similar functions in both species, making Amn a candidate gene for miscarriages in
humans. Amn may be required in the extra-embryonic endoderm during early embryonic
development, and in the intestine for the uptake of vitamin B12 in adults in both species.
Table 2. Comparison between mouse and human stages of early development and the timepoints at which
TM, AMN, and EPCR mutant mice embryos die. The carnegie stages (CS) provide a universal system for
staging and comparing the embryonic development of vertebrates. The stages are based on the external and
internal morphological development of the embryo, and are not directly dependent on either age or size. The
stages covers approximately the first 60 days of human development and nearly the entire intrauterine
development of mouse (UNSW Embryology, http://embryology.med.unsw.edu.au/embryo.htm) The table
shows the age of human and mouse embryos in days at specific stages. Implantation takes place during CS
5, gastrulation starts during CS 7, neurogenesis starts during CS 8, and heart beats can be detected at CS 10.
Developmental stages at
Carnegie stage Human Age Mouse Age lethality of mutations
CS 1 E 1 E 0.5
CS 2 E 2-3 E 1-2
CS 3 E 4 E 3
CS 4 E 5-6 E 4
CS 5 E 7-12 E 5-6 Implantation
CS 6 E 13-15 E 6.5
CS 7 E 15-17 E 7 Gastrulation
CS 8 E 17-19 E 7.5 Neurogenesis
CS 9 E 19-21 E 8
CS 10 E 22-23 E 8.5 Heart beats
CS 11 E 23-26 E 9 TM mutants die
CS 12 E 26-30 E 9.5
CS 13 E 28-32 E 10 AMN mutants die
CS 14 E 31-35 E 10.5 EPCR mutants die
CS 15 E 35-38 E 11
CS 16 E 37-42 E 11.5
CS 17 E 42-44 E 12
CS 18 E 44-48 E 13
CS 19 E 48-51 E 13.5
CS 20 E 51-53 E 14.5
CS 21 E 53-54 E 15
CS 22 E 54-56 E 15.5
CS 23 E 56-60 E 16
35
3.3.2 Thrombomodulin and Endothelial Protein C Receptor
The Thrombomodulin mouse model
In mice, absence of the blood-clotting regulator thrombomodulin (TM) causes embryonic
lethality before development of a functional cardiovascular system (Healy et al., 1995).
TM has been shown to have a dual role in development as TM is needed in two distinct
tissues. Expression of TM in non-endothelial cells of the placenta is required for proper
function of the early placenta, while the absence of TM from blood vessel endothelium
causes excessive activation of the embryonic blood coagulation system (Healy et al., 1995;
Isermann et al., 2001). The first indications of abnormal development are observed at E8.5
in homozygous TM-deficient (TM-/-) embryos and manifest as overall growth retardation.
This initial phenotype is invariably followed by rapid and complete resorption of
homozygous TM-null embryos within the next 10-12 hours (Table 2) (Healy et al., 1995;
Isermann et al., 2001).
Functions of the TM gene
TM is an endothelial cell surface glycoprotein expressed on the endothelium of arteries,
veins, capillaries, lymphatics, and on trophoblast cells of the placenta (Maruyama et al.,
1985; Weiler-Guettler et al., 1996; Fazel et al., 1998). TM plays a key role in the protein C
anticoagulant pathway, which is a major regulatory system providing an inhibitory
feedback mechanism that suppresses blood coagulation by preventing fibrin formation.
TM inhibits blood coagulation by forming a 1:1 complex with thrombin. Binding of
thrombin to this high-affinity receptor alters its specificity toward several substrates. The
complex converts protein C into the natural anticoagulant activated protein C (APC)
approximately 1000 times faster than thrombin alone. APC proteolytically inhibits the
coagulation cofactors Va and VIIIa, resulting in down-regulation of the activity of the
coagulation system (Figure 3) (Maruyama et al., 1985; Van de Wouwer et al., 2004;
Dahlbäck and Villoutreix, 2005).
A number of polymorphisms in the TM gene have been characterized and their
association with thrombosis or arteriosclerosis are subjects of ongoing research. These
associations have been evaluated in several studies and the results suggest that the
relevance of TM mutations in these complications is small or limited to a small subgroup
of individuals (Ireland et al., 1997; Doggen et al., 1998; Le Flem et al., 1999; Nakazawa et
al., 2002). The most convincing evidence that reduced TM function causes thrombosis is
36
derived from animal studies. In the past three genetically engineered mouse lines with various degrees of TM dysfunction have been generated, showing different degrees of thrombosis (Healy et al., 1995; Healy et al., 1998; Isermann et al., 2001). It appears that mutations causing a substantial reduction of TM function are very rare in humans (Faioni et al., 2002), possibly because such loss-of-function mutations would lead to foetal loss at an early developmental stage.
Figure 3. The protein C anticoagulant pathway. 1) Activation of prothrombin (PT); the Xa-Va complex converts prothrombin to thrombin (T), which forms a complex with TM. 2) Protein C activation; protein C (PC) binds to EPCR, which presents PC to the T-TM complex. 3) Cofactor inactivation; activated protein C (APC) binds protein S (PS). This complex inactivates factors Va and VIII thereby preventing blood coagulation. The EPCR mouse model Deletion of the endothelial protein C receptor (EPCR) gene in mice leads to embryonic lethality. Genotyping of progeny obtained from EPCR+/+/ interbreeding indicates that EPCREPCR / embryos die on or before E10.5 (Table 2). However, EPCRHowever, EPCR // embryos removed from extra-embryonic membranes and tissues at E7.5 and cultured in vitro develop beyond E10.5, suggesting a role for EPCR in the normal function of the placenta and/or at the maternal-embryonic interface (Gu et al., 2002). EPCR is normally detected on giant trophoblast cells, which are in direct contact with the maternal circulation and its clotting factors. In the giant trophoblast cells derived from EPCRcells derived from EPCR // embryos thrombosis is observed (Li et al., 2005). These observations provide evidence that extraembryonic EPCR expression is essential for embryonic viability and plays a critical role in the control of blood coagulation at the foeto-maternal boundary (Gu et al., 2002).
TM
37
Functions of the EPCR gene
EPCR is a type 1 transmembrane protein and is strongly expressed in the endothelial cells
of arteries and veins in heart and lung, and less intensely expressed in capillaries in the
lung and skin (Laszik et al., 1997; Simmonds and Lane, 1999; Esmon, 2000). It is also
expressed in the placenta and the developing cardiovascular system of the foetus (Crawley
et al., 2002). EPCR is seen as a complementary cofactor in protein C activation (Figure 3).
It binds both protein C and APC with similar affinity and functions in the protein C
pathway by binding protein C and presenting it to the TM-thrombin complex on the
endothelium, thereby further enhancing the rate of protein C activation and suppressing
blood coagulation (Stearns-Kurosawa et al., 1996; Van de Wouwer et al., 2004).
In human, sequence variations in the EPCR gene are associated with the risk of
thrombosis. A heterozygous mutation in EPCR has been identified in humans who
develop thrombosis and myocardial infarction. This mutation, c.323-9_336dup, is a 23-bp
insertion that leads to the formation of a truncated receptor that lacks the extracellular and
transmembrane domains and is unable to sustain protein C activation (Biguzzi et al.,
2001). However, a common polymorphism in EPCR, c.717+16G>C, seems to reduce the
risk of thrombosis. The CC genotype reduces the risk of venous thromboembolism 2.6-
fold, probably due to the higher APC levels observed in individuals carrying this genotype
(Medina et al., 2004; Medina et al., 2005).
TM and EPCR as candidate genes
Pregnancy complications in patients with prothrombotic risk factors have mainly been
linked to the formation of blood clots in placental blood vessels and sinuses, eventually
leading to intrauterine growth retardation and, in the most severe case, foetal loss
(Kupferminc, 2003; Kujovich, 2004; Kutteh and Triplett, 2006). Therefore, mutations in
the genes involved in the blood coagulation systems may increase the risk of thrombosis
and thereby miscarriage. Although the importance of the TM-protein C-EPCR system for
placental development and maintenance of pregnancy is firmly established in mice (Healy
et al., 1995; Gu et al., 2002), the relevance of these mechanisms for pregnancy associated
complications in humans remains to be established. The data from mouse models, the
known sequence homology of mouse and human TM and EPCR, and the similar type of
placentation in both species (Dittman and Majerus, 1989; Cross et al., 1994), however,
suggests TM and EPCR as candidate genes for RM in humans.
38
3.3.3 p53
Functions of the p53 gene
The p53 gene encodes a transcription factor consisting of an N-terminal transactivation
domain, a central sequence specific DNA binding domain, and a C-terminal
oligomerization domain (Levine, 1997). p53 is expressed at low levels in most cells and
the protein levels increase in response to various stress signals. p53 is expressed at high
levels in the mouse embryo up to midgestation, after which the expression levels rapidly
decrease (Rogel et al., 1985; Marzusch et al., 1995). Hundreds, if not thousands, of studies
have been dedicated to the functions of p53 because it plays a pivotal role in tumour
suppression. Indeed, more than 50% of human cancers contain somatic mutations in this
gene (Levine, 1997). p53 functions as a tumour suppressor by regulating the cell cycle,
protecting the genome from DNA damage and in inducing apoptosis. Moreover, p53
appears to interact with a large number of proteins involved in growth control, DNA repair
and transcriptional regulation (Choi and Donehower, 1999; Cheah and Looi, 2001; Balint
and Vousden, 2001). Experiments in mouse models have revealed that most of the human
tumour suppressor genes are essential for embryonic development. Furthermore,
inactivation of both germ line alleles of a tumour suppressor usually results in embryonic
lethality either in early or midgestation (Rogel et al., 1985; Schmid et al., 1991; Jacks,
1996), making p53 a candidate gene for RM.
There is one previous study supporting the hypothesis that p53 has a potential role
during early pregnancy also in humans. The results of Pietrowski et al. (2005) indicate that
RM is associated with a polymorphism 12139G>C at codon 72, resulting in either a
proline or arginine residue. In addition, the frequency of the Pro72 allele has been shown
to be significantly higher among women experiencing recurrent implantation failure than
in control women (Kay et al., 2006). The codon 72 variation of p53 have been shown to
differ biochemically and biologically (Thomas et al., 1999), but the role of p53 and the
codon 72 polymorphism in human development needs to be further studied.
p53 mouse models
Although p53 null mice can be viable and fertile, the absence of p53 activity can lead to
fatal developmental abnormalities and a significant proportion of p53-/- mice die during
embryogenesis. A substantial fraction of p53 null mice have developmental defects,
including neural tube defects, and the affected embryos frequently undergo resorption
39
clearly implicating p53 as an important factor in the early development (Sah et al., 1995;
Armstrong et al., 1995).
p53 can inhibit DNA replication by triggering growth arrest in G1 or initiate apoptosis,
and therefore its activity must be tightly regulated in order to allow normal growth. In the
absence of stress inducing signals, p53 is kept at low levels via its interaction with mouse
double minute 2 (MDM2) that binds to the transcriptional activation domain of p53. For
cell division in the early embryo to proceed, p53 needs to be down-regulated by Mdm2
(Piette et al., 1997; Bond et al., 2005). A deletion of Mdm2 causes embryonic lethality in
mice around the time of implantation (Montes de Oca Luna et al., 1995; Jones et al.,
1995). In addition, p53QS
mutant mice, which cannot bind Mdm2 due to mutations in
residues crucial for transactivation, show increased embryonic lethality in a homozygous
state (Johnson et al., 2005).
Mouse models have also suggested that p53 may play a key role in development
through protecting embryos from teratogens and diverse environmental stresses (Nicol et
al., 1995; Norimura et al., 1996). Teratogenesis experiments in mice provide strong
support for the idea that p53 efficiently aborts embryonic cells with teratogen induced
DNA damage. If too many cells of the embryo die through apoptosis, the entire embryo
will die. Conversely, lack of p53 in the embryo presumably results in continued cell cycle
progression despite DNA damage, much less apoptosis and a higher proportion of
embryos with developmental anomalies (Nicol et al., 1995). Taken together, the results
obtained from mouse models suggest that the relative levels of p53 in the early embryo are
highly important, and the protein expression has to be very carefully controlled during
developmental processes. Changes in the expression levels of p53 elevates the risk of
miscarriage and developmental defects, and increases the susceptibility to chemical
teratogenesis, possibly also in humans.
3.3.4 The mitochondrial genome
Mitochondria are cellular organelles that contain their own genome. It is estimated that in
a somatic cell one human mitochondrion typically contains 5-10 DNA copies and,
depending on the cell type, a somatic cell can contain up to 1,000 mitochondria. The
human mitochondrial genome is circular, double-stranded and 16,569 bp in size. It
40
contains 37 genes; 13 protein coding genes, which are all components of the respiratory
chain; two rRNAs; and 22 tRNAs needed for mitochondrial protein synthesis (Howell et
al., 2000; DiMauro, 2001; Thorburn and Dahl, 2001). The mitochondria are involved in
cellular ATP production and apoptosis (DiMauro, 2007). These processes are of great
importance in the early development of embryos and may be disturbed by mutations in
mitochondrial DNA (mtDNA).
Genetics of mtDNA
Mitochondrial genetics differs from Mendelian genetics in three major aspects:
1) Maternal inheritance. At fertilization, all mitochondria in the zygote derive from the
oocyte. mtDNA mutations are therefore inherited in a matrilineal fashion (DiMauro, 2001;
McKenzie et al., 2004).
2) Heteroplasmy/threshold effect. In contrast to autosomal nuclear genes, each
consisting of one maternal and one paternal allele, there are hundreds or thousands of
mtDNA molecules in each cell. Deleterious mutations in mtDNA usually affect some but
not all of the mtDNA genomes. Therefore cells, tissues, and organs can contain two
different mtDNA populations (wild-type and mutant), a state known as heteroplasmy. The
situation when all mtDNAs are identical is called homoplasmy. Non-deleterious
mutations/neutral polymorphisms of mtDNA are usually homoplasmic, whereas
pathogenic mutations are usually heteroplasmic. For an mtDNA mutation to have a
deleterious effect on tissue function it must reach a minimum threshold - a certain
proportion of mutant molecules are required before a phenotype is observed (DiMauro,
2001; Thorburn and Dahl, 2001; McKenzie et al., 2004).
3) Segregation. As a heteroplasmic cell replicates different proportions of mutant
mtDNA molecules are transmitted to the daughter cells, and when the pathogenic
threshold for a given tissue is surpassed the phenotype can also change. In this way the
mtDNA genotype can drift toward homoplasmy, either wild-type or mutant (DiMauro,
2001; McKenzie et al., 2004).
Today over 200 mtDNA point mutations have been reported in the Mitomap database
(http://www.mitomap.org) to be associated with different human diseases. These
mutations can be divided into two groups: mutations affecting mitochondrial protein
synthesis, including mutations in tRNA and rRNA genes and mutations in protein
encoding genes. The mutations are involved in a broad spectrum of human diseases,
41
including multisystem disorders as well as more tissue specific diseases (DiMauro, 2001;
Taylor and Turnbull, 2005; McKenzie et al., 2004; Wong, 2007).
The mitochondrial bottleneck
The segregation of mtDNA alleles is unexpectedly rapid, and in some maternal lineages
complete allele switching has been observed in a single generation without a
heteroplasmic intermediate (Koehler et al, 1991). Rapid changes in mitochondrial allele
frequencies were first observed in cattle (Olivo et al., 1983; Ashley et al., 1989) but
similar changes have subsequently been described in many mammalian species, including
humans. Analysis of human pedigrees segregating pathogenic mtDNA point mutations
show large differences in the proportion of mutant mtDNAs among siblings and between
generations (Larsson et al., 1992; Ghosh et al., 1996; Jacobi et al., 2001).
The rapid changes in mtDNA allele frequency have been explained by an mtDNA
bottleneck in the germ line, responsible for the random segregation of mtDNA sequence
variations between generations. A mature oocyte contains about 100,000-200,000 copies
of mtDNA (Figure 4). These mtDNA copies are randomly divided, without mtDNA
replication, into cells of the developing embryo. Replication of mtDNA is initiated when
the primordial germ cells (PGC) migrate and differentiate to generate oocytes. It is
estimated that human PGCs contain ~10-100 mitochondria, increasing to ~200-1000 in the
oogonia and to several thousands in the oocyte in the primordial follicle. As the oocyte
matures the number of mitochondria again increases to >100,000 (Jenuth et al., 1996;
Shoubridge, 2000; Khrapko, 2008; Stewart et al., 2008). Segregation is rapid because the
precursor cells contain relatively small numbers of mtDNA templates and because the
replication of mtDNA is under relaxed control. This relaxed form of replication allows
some templates to replicate more than once during the cell cycle, while others are not
replicating at all. Because of the mitochondrial bottleneck, the mtDNA genotype in the
embryo is largely determined by the genotype of the few mitochondria in the PGC in the
mother (Jenuth et al., 1996; Shoubridge, 2000).
mtDNA in development
The mitochondria produce the energy needed for cellular processes by producing ATP
through oxidative phosphorylation. Through dysfunctions of energy metabolism or
regulation of apoptosis, the mitochondria are suggested to contribute to human oocyte
wastage and early embryo demise (Van Blerkom, 2004). The role of mtDNA mutations in
42
human early development is supported by a report describing a family with an unusual
homoplasmic tRNA mutation that resulted in six neonatal deaths and one surviving child
with Leigh syndrome even though the mother is clinically unaffected (McFarland et al.,
2002). As the proportion of mutant mtDNA copies can shift when inherited the phenotype
often shows remarkable variation within a family (Ghosh et al., 1996; Blok et al., 1997;
Jacobi et al., 2001). Accordingly, such a shift can occur from a mother with mild or no
phenotype to an embryo with significant enrichment of mutated mtDNA causing foetal
loss and possibly RM. Homoplasmy for a severe pathogenic mtDNA mutation is rarely
observed, presumably because such a condition would be lethal. For example mutations
causing mitochondrial encephalopathy with lactic acidosis and stroke-like episodes
(MELAS) and myoclonic epilepsy and ragged-red fibres (MERRF) are invariably
heteroplasmic and homoplasmic mutations are presumed to be fatal (Jenuth et al., 1996;
Howell et al., 2000). An oocyte with such a mutation in a homoplasmic state would most
likely result in foetal death at an early developmental stage. It has been suggested that this
phenomenon might even be beneficial, if spreading of the deleterious mtDNA mutations
that slip through the bottleneck would thus be eliminated (Bergstrom and Pritchard, 1998;
Howell et al., 2000).
Figure 4. The mitochondrial bottleneck. During oocyte maturation a limited number of mtDNA copies are
transferred to each primary oocyte. As this small population of mtDNA copies is rapidly replicated it may
lead to a random shift of the mutational load in the next generation. Due to the mitochondrial bottleneck the
offspring of a heteroplasmic female may receive a varying number of mutant mtDNA copies at fertilization
resulting in different phenotypes; A) low level of mutation (unaffected offspring), B) intermediate level of
mutation, C) high level of mutation (affected offspring). Modified from Taylor and Turnbull, 2005.
43
Aims
The present study is based on hypotheses that some as yet unknown genetic factors may
result in RM. Consequently, the main aim of this study was to gain new information about
the underlying genetic causes of RM in the Finnish population. The aim was to identify
candidate genes for RM and subsequently screen these genes for mutations in a series of
samples with RM. In addition, we aimed to study if genetic factors previously reported to
underlie RM could be the cause of RM in Finnish patients.
The following specific aims were set for the study:
1. To determine whether ancestral places of origin of the couples with RM show
geographic clustering.
2. To search for mutations in four candidate genes for RM ─ Amnionless,
Thrombomodulin, Endothelial Protein C Receptor and p53 ─ identified to be cruical for
development in mouse models.
3. To study the expression of p53 in placental samples.
4. To search for variations associated with RM in the mitochondrial genome.
5. To study if specific sex chromosome characteristics, namely skewed X chromosome
inactivation or Y chromosome microdeletions, are associated with RM in Finnish patients.
44
Materials and methods
1 Study subjects
1.1 Patients (I, II, III, IV, V)
The patients recruited for this study have been treated for unexplained RM at the
Department of Gynaecology and Obstetrics of the Helsinki University Hospital during the
years 2001-2004. The inclusion criteria for the 48 women included in the study were:
1) age 18-40 years (mean age 31 years)
2) previous history of RM, defined as three or more consecutive miscarriages
3) normal karyotype
4) no uterine abnormalities (examined by ultrasonography or hysterosonogram)
Blood samples were obtained from male partners of 40 of the women included in the
study. The karyotypes of these males were checked and found to be normal. Consequently,
a total of 40 couples and 8 women with unexplained RM were included in the study. The
women in the study had experienced three to seven miscarriages. In 45 women all
miscarriages had taken place during the first trimester (weeks 1-12 of pregnancy). Two
women had experienced a second trimester (weeks 13-26 of pregnancy) miscarriage, and
one women a third trimester (from week 27 to birth) intrauterine foetal death in addition to
first trimester losses.
1.2 Controls (I, II, III, IV, V)
The control group consisted of 191 healthy women who had had at least one normal
pregnancy and no known history of miscarriage. They were recruited from the same
hospital during the same time period as the patients. The controls were of the same age
and racial origin (Finnish Caucasian) as the patients.
45
1.3 Placental samples (III, IV)
One or more paraffin embedded placental sample from a spontaneously aborted pregnancy
were available from seven of the RM women included in the study. Altogether, placental
samples were available from 19 miscarriages.
1.4 Ethical issues (I, II, III, IV, V)
Informed consent was obtained from all participants prior to enrolment in the study. The
ethics committee of the Department of Obstetrics and Gynaecology, Helsinki University
Central Hospital, has approved this study.
2 Methods
2.1 Genealogic analysis (I)
Data concerning the birthplaces of the patients´ parents and grandparents were collected to
determine the ancestral places of origin for the patients and to investigate if the birthplaces
were clustered to specific regions of Finland. Data concerning the birthplaces were
obtained through a questionnaire. Replies were received from 82 (95%) patients. To
investigate the geographic distribution of ancestral birthplaces, the birthplaces were placed
onto a map of Finland, including the formerly Finnish areas of Russian Karelia and
compared to the population distribution of Finland.
2.2 DNA extraction (I, III)
DNA was extracted according to the manufacturer´s instructions from whole blood
samples and paraffin sections using Puregene DNA isolation kits (Gentra Systems,
Minneapolis, USA).
46
2.3 Candidate gene analysis
Polymerase chain reaction and dHPLC sample analysis (I, II, III, IV)
The exons and exon-intron boundaries of the candidate genes were amplified using
polymerase chain reactions (PCR). Amplification of appropriately sized PCR products
were confirmed by agarose gel electrophoresis before further analysis. Mutation screening
of the candidate genes was mostly carried out using denaturing high performance liquid
chromatography (dHPLC) with the Transgenomic WAVE Nucleic Acid Fragment
Analysis System (Transgenomic, Omaha, USA) and the associated Navigator software. To
obtain optimal resolution of homoduplex and heteroduplex DNA fragments the
temperature was set for partially denaturing conditions. The melting profile of the
amplicons and the optimal oven temperature was predicted for each PCR fragment using
the Navigator software. The buffer gradients for the elution of the fragments were created
automatically with the Navigator software.
To allow sequence analysis of low-level heteroplasmic mtDNA, fractions of small
sized heteroduplex peaks were collected using an FCW-200 fraction collector integrated
with the WAVE system. To increase the heteroplasmy level in the samples the DNA
fractions were reamplified with the primers used to produce the original amplicon.
Sequencing (I, II, III, IV)
Following dHPLC analysis, samples showing heterozygous peaks were sequenced in order
to determine the nature of the sequence change. PCR products were purified and
sequenced using BigDye version 3.1 sequencing chemistry and an ABI 3730 DNA
Analyzer (Applied Biosystems, Foster City, USA) according to the manufacturer's
instructions.
Genotype determination (I, II, III)
Variations that were detected in a homozygous state by sequencing were genotyped by
restriction enzymes, sequencing, allele specific PCR, or dHPLC.
Solid-phase minisequencing (IV)
Quantification of heteroplasmy levels were performed using solid-phase minisequencing
according to a previously described protocol (Suomalainen and Syvänen, 2000). PCR
47
products were amplified using specific primers designed for each of the variations studied.
Biotinylated PCR products were then captured in streptavidin-coated microtiter wells and
single nucleotide extension of specific detection primers were performed using
radioactively labelled nucleotides. The results were expressed as numeric cpm-values,
where the incorporated label expressed the relative amount of sequence variation in the
sample.
Quantitative real-time PCR (IV)
The mitochondrial cytochrome b (Cytb) and the nuclear amyloid beta precursor protein
(APP) genes were simultaneously amplified by quantitative TaqMan real-time PCR assay
in an ABI Prism 7000 Detection System Cycler to quantify the amount of mtDNA.
Fluorescence acquisition was done at the end of each annealing step, and the relative
levels of the genes were compared in the exponential phase of PCR, using the ABI Prism
7000 SDS software (Applied Biosystems). The ratio of mtDNA to nuclear DNA was used
as a measure of mtDNA content (2-ΔΔC
T method) (Livak and Schmittgen, 2001).
Predicting the effects of variations (I, II, III, IV)
The ESEfinder (http://exon.cshl.edu/ESE/) is a web based resource which scans nucleotide
sequences to identify putative exonic splicing enhancers (ESEs), which act as binding sites
for splicing factors. The ESEfinder was used to predict if the exonic or intronic sequence
variations disrupt the sequence of known ESE-elements. The ESEfinder searches for
putative ESE motifs and calculates a score for all sequences with a motif match. A
sequence is considered to be potentially needed for splicing when the calculated score is
greater than the threshold value defined by the program.
The variations predicted to change an amino acid were analyzed by the SIFT (sorting
intolerant from tolerant) program (http://blocks.fhcrc.org/sift/SIFT.html) which predicts
whether an amino acid substitution in a protein will be tolerated.
Different web based databases were used to determine if the detected variations have
been previously reported; Ensemble Genome browser (http://www.ensembl.org), UCSC
Genome browser (http://genome.ucsc.edu), NCBI Entrez SNP database
(http://www.ncbi.nlm.nih.gov), p53 knowledgebase (http://p53.bii.a-star.edu.sg),
Mitomap (http://www.mitomap.org), and mtDB-Human Mitochondrial Genome Database
(http://www.genpat.uu.se/mtDB/).
48
2.4 Immunohistochemical analysis of p53 (III)
5μm-thick sections were cut from paraffin blocks containing placental samples. The
sections were deparaffinized in xylene and rehydrated through graded concentrations of
ethanol to distilled water. Monoclonal antibody DO-7 directed against p53 protein
(dilution 1:100; Dako, Glostrup, Denmark) was used as primary antibody (incubation time
55 min). The procedure was performed in a TechMate 500 automated machine (DAB
detection kit; Dako ChemMate). The sections were lightly counterstained with Mayer's
hematoxylin.
2.5 X chromosome inactivation analysis (V)
The XCI ratio was determined in the females by analysing the methylation status of the X-
linked androgen receptor (AR). Exon 1 of the AR gene contains a polymorphic
microsatellite with varying numbers of trinucleotide repeats (CAG)n which can be used to
identify the different alleles of the locus. DNA was digested with a methylation sensitive
restriction enzyme that digests only nonmethylated (active X) DNA segments. Methylated
sites can not be digested and remain intact for amplification.
Fragment analysis of the microsatellite was performed using an ABI 3730 DNA
Analyzer (Applied Biosystems). Fragment analysis data were analysed and genotyped
using GeneMapper 3.0 software. The total fluorescent peak areas for both alleles in
digested and undigested samples were recorded and used for calculation of the ratio of
allele inactivation. Previously described equations (Lau et al., 1997; Iitsuka et al., 2001)
were used to calculate the degree of inactivation. We used both 85% (mildly skewed) and
90% (extremely skewed) inactivation of a particular X chromosome as cut off points for
skewed X-inactivation.
2.6 Y chromosome microdeletion analysis (V)
Males were screened for the presence of microdeletions using Y chromosome specific
sequence tagged sites (STS) primer sets. Altogether, 37 STS loci spanning the whole Y
chromosome were analysed in multiplex PCR reactions. The presence of the specific Y
chromosome markers were shown by observing the PCR fragments on agarose gels.
49
2.7 Statistical analysis (I, II, III, IV, V)
X 2 and Fisher´s exact tests ([SISA 1997 programs] http://home.clara.net/sisa) were used
for statistical analyses of the data. The odds ratios (OR) and 95% confidence intervals (CI)
of the OR were calculated to assess the relative risk conferred by a particular allele or
genotype. Fisher´s exact test was also used to compare the frequency of skewed XCI in the
patient and control groups. The T-test was used to calculate differences in mtDNA copy
number levels after real-time PCR analyses. Differences were considered as statistically
significant for p-values <0.05.
50
Results and discussion
The fact that the observed incidence of RM is much higher than expected by chance alone
suggests that RM is a clinical entity distinct from sporadic miscarriages (Regan, 1991).
This is supported by the finding that RM tends to occur, unlike sporadic miscarriage,
when the foetus has a normal chromosome constitution (Stirrat, 1990; Sullivan et al.,
2004). Recently, the genetic knowledge about the aetiology of diseases has expanded
rapidly. Progress has, however, been much slower in elucidating the genetic causes of
female infertility and RM. Although many studies have been conducted to identify the
underlying mechanisms, the cause of miscarriage remains unknown in approximately 50%
of cases (Plouffe et al., 1992; Clifford et al., 1994). Consequently, this study was
conducted to gain further information about the possible genetic causes of RM in our
Finnish patients.
1 Genealogic studies (I)
Genealogical studies can be used to identify geographic enrichment of genetic disorders.
Depending on the age of the founder mutation the geographical area of such an enrichment
can be very restricted or include large regions. In many diseases of the Finnish disease
heritage a regional clustering of cases can still be observed, reflecting the geographical
origins of ancestral birthplaces (Norio, 2003a; Norio, 2003b). For example, a specific type
of ovarian failure, caused by mutations in the follicle stimulating hormone receptor
(FSHR), called follicle stimulating hormone-resistant ovaries (FSH-RO) is an autosomal
recessive trait which shows geographical enrichment of ancestral birthplaces (Aittomäki et
al., 1995; Aittomäki et al., 1996). We conducted genealogic studies to investigate if such
an enrichment reflecting a common genetic cause could underlie the miscarriages in our
patients. The ancestors of all couples with RM were traced by questionnaires enquiring as
to the full names, dates and places of birth of the patients´ parents and grandparents.
Replies were received from 82 patients. To investigate the geographic distribution of the
birthplaces, the birthplaces of the parents and grandparents were placed onto a map of
Finland (Figure 5) and compared to a reference map showing the distribution of the
Finnish population.
51
The results of the genealogic studies suggest the possibility of a shared underlying
genetic defect for the miscarriages in a subset of patients. Compared to the reference map
of the general population distribution in Finland the geographical distribution of the
birthplaces of the parents and grandparents of the patients is uneven (Figure 5). The results
suggest that there may be an eastern enrichment in the province of Kuopio of the ancestral
birthplaces. In addition, the Helsinki region is over-represented, but this is explained by
the fact that the patients were all collected from Helsinki. The clustering around Kuopio
suggests that some of the patients may, due to common ancestors, carry a mutation
increasing the susceptibility to RM. This would not be exceptional in Finland due to the
population history, which has increased the local incidence of rare disorders. The special
assortment of rare diseases among Finns is caused by the fact that the population living in
Finland today is descended from a relatively small original population. As the population
has increased, some alleles increased in frequency while others disappeared due to genetic
drift, population bottlenecks, and non-assortative mating (Norio, 2003a; Norio, 2003b).
Figure 5. Birthplaces of the A) parents and B) grandparents of 82 patients with RM. 34 of the parents and
24 of the grandparents were born in Helsinki. The population density of Finland at the time the parents were
born is shown in the small reference map.
A) B)
52
2 Candidate gene analysis (I, II, III, IV)
The results of the genealogic studies suggest a common underlying genetic cause for RM
in at least a subgroup of our patients. Therefore, we screened a number of candidate genes
for mutations that could possibly explain the RM experienced by the couples. The
candidate genes AMN, TM, EPCR, and p53 were selected based on findings in animal
models. When screening candidate genes in which homozygous mutations are expected to
be lethal in early pregnancy, the affected foetuses cannot, in most cases, be studied
because DNA from the spontaneously aborted foetuses is rarely available. Such mutations
can be studied using placental tissue samples or by screening couples experiencing RM, in
which case partners are expected to be healthy heterozygous carriers of mutations in the
candidate gene of interest. In addition to candidate genes identified from animal models,
the genes of the maternally inherited mitochondrial genome were chosen as candidate
genes based on their known functions.
2.1 Amnionless (I)
Due to the fact that homozygous mutations in Amn are known to cause miscarriage in the
mouse (Wang et al., 1996) and the human AMN and mouse Amn genes are highly
homologous (Kalantry et al., 2001) AMN was considered a candidate gene for
miscarriages in humans. Therefore, we examined whether there is an association between
RM and sequence variations of the AMN gene in humans. To our knowledge this is the
first study concerning the role of AMN in miscarriage.
2.1.1 Variations in the Amnionless gene
A total of 14 sequence variations were found by screening 85 patients (40 couples and 5
women) with a history of unexplained RM and 95 controls for mutations in the 12 exons
and the untranslated regions (UTR) of the Amnionless gene. Eleven of the detected
sequence variations are intronic. Of these variations, 10 are single nucleotide substitutions,
and one variation is a 10 basepair (GCGTGGCGTG) duplication. Three of the sequence
variations are exonic, of which two are single nucleotide substitutions and one is a
duplication (Table 3). None of the exonic variations detected in AMN had been previously
53
reported (databases used: www.ncbi.nlm.nih.gov; www.ensembl.org). In total, 8 novel
sequence variations were detected in the AMN gene (Table 3).
The sequence variation c.363G>A is a single nucleotide substitution in exon 5,
changing the last nucleotide of codon 121 (GGG->GGA). It is a synonymous variation,
detected only in patient samples and does not predict a change of the amino acid (glycine).
Sequence variation c.829A>G is a non-synonymous SNP in exon 8, changing the first
nucleotide of codon 276 (ACC->GCC). The variation is predicted to change the amino
acid from threonine to alanine. According to the SIFT program this amino acid change
would be tolerated even though the hydroxyl group of threonine makes it much more
hydrophilic and reactive than alanine. Sequence variation c.1339_1344dup is a duplication
in exon 12. It is a 6 basepair duplication (GCCGGG) of the codons 447 and 448, located 5
codons before the stop codon. This DNA sequence change is predicted to make the final
protein product 2 amino acids (Ala+Gly) longer.
Table 3. The location of the variations in the AMN gene, predicted amino acid changes, and the numbers of
patients and controls carrying the rarer allele in a hetero/(homozygous) state. Numbering of the nucleotides
is relative to the adenine in the ATG start codon in the reference sequence (UCSC Genome Browser,
http://genome.ucsc.edu/).
AMN sequence variations
RM women+
DNA variation male partners RM women controls
(predicted amino acid change) Location n=85 n=45 n=95
c.-87C>G Us 5´UTR 13 9 15
c.-74C>T Us 5´UTR 10 8 11
c.-27T>C Us 5´UTR 48 (15) 24(9) 50 (19)
c.-23G>C Us 5´UTR 46 (9) 22(5) 44 (9)
c.296-75_-66dup 1 intron 4 0 0 1
c.363G>A 1 exon 5 4 4* 0
c.829A>G (Thr276Ala) 1 exon 8 1 1 1
c.843+11C>T 1 intron 8 5 5* 1
c.1169+42C>G intron 10 43 (12) 27(6) 47 (14)
c.1170-6C>T 1 intron 10 2 1 2
c.1339_1344dup 1 exon 12 2 0 2
c.1362+38G>C 3´UTR 34 (11) 22(3) 40 (14)
c.1362+518C>T 1 Ds 3´UTR 0 0 1
c.1362+523G>A 1 Ds 3´UTR 0 0 1
1 Novel variation
Us=upstream of, Ds=downstream of
* p < 0.05
54
Two of the novel variations; c.363G>A in exon 5, and c.843+11C>T in intron 8 were
significantly more frequent in RM women compared to control women, indicating that
these variations may increase the risk of miscarriage. The c.363G>A variation was
detected in 4 RM women (8.7%) and in none of the control women (p=0.010). The
c.843+11C>T variation was detected in 5 RM women (10.9%) and in one of the control
women (1%) (p=0.014).
2.2 Thrombomodulin and Endothelial Protein C Receptor (II, unpublished)
Mutations in the TM or EPCR genes may cause thrombophilia in the mother, and thereby
constitute a risk factor for miscarriages. In addition, homozygous mutations in the foetus
may cause miscarriage by other mechanisms as well (Isermann et al., 2001). An important
role for the TM-protein C-EPCR system for placental development and maintenance of
pregnancy has been shown in mouse models (Healy et al., 1995; Gu et al., 2002). The
relevance of these mechanisms for pregnancy associated complications in humans has,
however, not yet been determined.
2.2.1 Variations in the TM gene
In this study we analysed the entire coding region of TM including exon-intron boundaries
and the 5´ and 3´ UTRs for mutations in 86 patients with RM and 191 controls. As a
result, one exonic and one 3´UTR sequence variation in the TM gene were detected (Table
4). These variations were detected in both patients and controls. There were no significant
differences in the allele or genotype frequencies between all RM patients or RM women
and controls for either of the sequence variations in the TM gene.
A novel 18 bp deletion in the 3´UTR of the TM gene, 1728+23_+40del, was detected
in five patients (6%) and five controls (3%). The variation is located 23 bp downstream of
the exon-intron boundary. The other variation detected in TM, c.1418C>T, is a previously
reported common polymorphism resulting in an amino acid substitution from an alanine to
a valine at codon 455. The variation is located in the sixth epidermal growth factor like
domain of the gene that is responsible for thrombin binding and thereby necessary for
protein C activation (Esmon, 1989) but has, however, previously been shown to be neutral
in respect to thrombophilia (van der Velden et al., 1991). The variation is likely to be
55
tolerated according to the SIFT program, and was commonly detected in a homozygous
state in both patients and controls in our study.
Table 4. The location of the variations in the TM gene, predicted amino acid changes, and the numbers of
patients and controls carrying the rarer allele in a hetero/(homozygous) state. Numbering of the nucleotides
is relative to the adenine in the ATG start codon in the reference sequence (UCSC Genome Browser,
http://genome.ucsc.edu/).
TM sequence variations
RM women+
DNA variation male partners RM women controls
(predicted amino acid change) Location n=86 n=46 n=191
c.1418C>T (Ala455Val) exon 1 41 (5) 26(3) 84 (8)
c.1728+23_+40del1 3´UTR 5 3 5
1 Novel variation
2.2.2 Variations in the EPCR gene
As a result of screening 86 patients and 191 controls for mutations in EPCR two exonic
variations and two variations in the non-coding regions of the gene were detected (Table
5). The two non-coding substitutions detected in EPCR have been previously reported as
common polymorphisms. Variation c.323-20T>C is located 20 bp upstream of exon 3 and
variation c.717+16G>C is located in the 3´ UTR. The c.717+16G>C polymorphism has
been suggested to have a modifier effect on the risk of venous thromboembolism (VTE).
Individuals homozygous for the C allele have elevated levels of activated protein C and a
lower risk of VTE (Medina et al., 2004). In addition, a recent study suggests that the C
allele decreases the risk of RM in women with a factor V Leiden mutation (Hopmeier et
al., 2008).
Variation c.655A>G is located in exon 4. It is a non-synonymous change, predicting a
Ser219Gly change. The variation is predicted to be tolerated according to the SIFT
program and previous studies have concluded that this variation does not increase the risk
of VTE (Medina et al., 2004; Ireland et al., 2005). The 23 bp duplication, c.323-9_336dup
in EPCR exon 3, duplicates the preceding 23 bases and results in a premature STOP codon
downstream of the insertion point. The mutation leads to the formation of a truncated
receptor lacking the extracellular and transmembrane domains. The truncated protein does
not function properly because it is not localized on the cell surface, cannot be secreted
from the cells, and does not bind activated protein C (Biguzzi et al., 2001). We detected
56
this variation in three samples; in one female patient, in one male partner, and in one
control woman. All four variations in EPCR were detected in both patients and controls.
There were no significant differences in the allele or genotype frequencies between all RM
patients or RM women and controls for any of the EPCR sequence variations. Based on
these results we cannot conclude a role for the detected EPCR variations in RM.
Table 5. The location of the variations in the EPCR gene, predicted amino acid changes, and the numbers of
patients and controls carrying the rarer allele in a hetero/(homozygous) state. Numbering of the nucleotides
is relative to the adenine in the ATG start codon in the reference sequence (UCSC Genome Browser,
http://genome.ucsc.edu/).
EPCR sequence variations
RM women+
DNA variation male partners RM women controls
(predicted amino acid change) Location n=86 n=46 n=191
c.323-20T>C intron 2 41 (15) 18(11) 95 (33)
c.323-9_336dup exon 3 2 1 1
(TyrProGlnPheLeuSTOP)
c.655A>G (Ser219Gly) exon 4 22 (1) 12 49 (2)
c.717+16G>C 3´UTR 41 (15) 19(11) 93 (34)
2.2.3 TM/EPCR variations and outcome of heparin treatment
Of the women included in this study, 42 had been treated with heparin and/or aspirin in
attempts to improve the pregnancy outcome. Many of the RM women in our study had a
full term pregnancy when treated with heparin, but not all (Ulander V-M, personal
communication). As mutations in the TM or EPCR genes may cause disturbances of the
uteroplacental vasculature, we investigated whether TM or EPCR variations could be
associated with pregnancy outcome after heparin/aspirin treatment. Unfortunately, the
number of patients in each group was too small to obtain statistically significant results
and therefore associations between a specific TM/EPCR variation and the outcome of the
heparin treatment could not be studied.
57
2.3 p53 (III, unpublished)
p53 functions have proven important for normal development in mice. However, its role in
human embryonic development has not been studied enough to make conclusions about its
role at different developmental stages. As the level of p53 expression and the ability of
p53 to interact with other proteins can be disturbed by sequence variations within the
gene, we screened the p53 gene for variations in our series of RM patients.
2.3.1 Variations in the p53 gene
Nine different sequence variations were found in the p53 gene by screening 40 couples
and 6 women with RM and 191 controls using dHPLC (Table 6). Six variations were
confirmed by sequencing to be located in the non-coding regions, and three variations
were found within the coding regions of the gene. All the variations have been previously
reported as common polymorphisms (p53 knowledgebase; NCBI SNP database). Two of
the exonic variations; G12032A, A13399G are synonymous and do not cause a change of
an amino acid, while one of the exonic variations, G12139C, is a common non-
synonymous polymorphism changing the amino acid of codon 72 from an arginine to a
proline. These R72P variants have been shown to differ biologically and biochemically
(Thomas et al., 1999), and have differences in their ability to interact with basic elements
of the transcriptional machinery, to induce apoptosis and to suppress transformed cell
growth (Dumont et al., 2003; Pim and Banks, 2004). This variation has previously been
studied with respect to RM. Pietrowski et al. (2005) studied 175 RM women and found an
association between the R72P polymorphism and RM, while Coulam et al. (2006) studied
205 RM women and showed no significant differences in the frequencies of R72P
polymorphism between RM patients and controls. In this study, we did not detect any
significant association between the R72P polymorphism and RM.
We found the polymorphism C11992A, located 29 bp upstream of exon 4, to be
associated with RM. The C/A and A/A genotypes of the C11992A polymorphism were
shown to be significantly more frequent (p-value <0.05) in women with RM than control
women. Women carrying the C/A or A/A genotype had a more than twofold increased risk
for miscarriage (p=0.0414, OR 2.083, CI 1.018-4.259). Statistical calculations using the
X2-test were performed following a dominant genotype model, where the C/C genotypes
58
were compared against the C/A and A/A genotypes. A C/A or A/A genotype was detected
in 32.6% of the women with miscarriages and in 18.9% of the controls. When comparing
the allele frequencies of the C11992A variation, there was no significant difference
between the patient and control groups.
Table 6. The location of the variations in the p53 gene, the predicted amino acid change, and the number of
patients and controls carrying the rarer allele in a hetero/(homozygous) state. Numbering of the nucleotides
and codons is relative to the sequence presented by the p53 knowledgebase (http://p53.bii.a-star.edu.sg)
2005.
p53 sequence variations
RM women+
DNA variation male partners RM women controls
(predicted amino acid change) Location n=86 n=46 n=191
11827 C>G intron 2 23 (6) 12(4) 42 (9)
11992 C>A intron 3 22 (2) 13(2)* 29 (7)
12032 G>A exon 4 2 0 1
12139 G>C (Arg72Pro) exon 4 38 (8) 22(3) 77 (8)
13399 A>G exon 6 2 1 0
13964 G>C intron 6 0 0 2
14181 C>T intron 7 5 3 16
14201 T>G intron 7 5 3 16
14766 T>C intron 9 1 0 2
* p < 0.05
2.3.2 p53 expression in RM placenta
Previous studies have shown that complete absence of p53 in mice can result in lethality,
reduced fertility, and high susceptibility to the effects of environmental stress. In addition,
deviations in expression levels of p53 can have serious detrimental consequences (Sah et
al., 1995; Nicol et al., 1995; Godley et al., 1996). Futhermore, a previous study showed
higher expression levels of apoptosis related genes in chorionic villi from RM patients
compared to controls (Choi et al., 2003). To further study the role of p53 in human
development, we studied the expression of p53 in placental tissue samples from
miscarriages of RM patients.
Placental tissue samples were available from 19 miscarriages from seven women.
These were subjected to immunohistological staining to study the expression of p53 in the
placenta. No abnormal expression patterns were detected in these samples (Figure 6). All
59
samples showed heterogenic and primarily weak p53 expression. Stronger expression was
mainly seen in the cytotrophoblasts and syncytiothrophoblasts.
Ten of the placental tissue samples were obtained from couples who both had a C/C
genotype at position 11992, and 9 of the samples were from couples where one partner
was a C/A heterozygous and the other a C/C homozygote. According to parental
genotypes none of the pregnancies from which placental tissue were obtained were A/A
homozygotes and hence we were not able to study whether the 11992A/A genotype would
affect the expression of p53 in the placenta.
Figure 6. A representative p53 staining of placental tissue obtained from a miscarriage.The expression
patterns were similar in all 19 tissue samples. Stronger heterogeneous p53 expression is seen in the
cytotrophoblasts and syncytiotrophoblasts.
2.4 Mitochondrial genes (IV)
Recent studies suggest that mitochondrial functions in the oocyte may have a critical role
in human embryonic development. Dysfunctions that affect the capacity of the
mitochondria to produce ATP or regulate the apoptosis cascade have been suggested to
cause early demise of the embryo in humans (Van Blerkom et al., 1998; Van Blerkom,
2004). Accordingly, we hypothesized that some women with RM could be healthy carriers
of low-level heteroplasmic mtDNA mutations which would predispose to RM.
60
2.4.1 Variations in the mitochondrial genome
In this study we screened 48 RM women and 48 age matched control women for
heteroplasmic mtDNA variations using dHPLC. The method is based on detection of
heteroduplexes in a sample and can therefore be used to detect heteroplasmic sequence
variations. However, by sequencing a number of homoplasmic deviations from the
reference sequence were also detected. Because all samples were not sequenced, we could
not assess the frequency of homoplasmic changes.
Table 7. The location of the variations in the mtDNA, predicted amino acid changes, and the numbers of
RM women and controls carrying the variation in a heteroplasmic state. Numbering of the nucleotides are
based on the reference sequence in the Mitomap database.
mtDNA sequence variations
DNA variation RM control
(predicted amino acid change) Location women women
16362 T>C non-coding 1 0
16519 T>C non-coding 1 1
73 A>G non-coding 1 1
146 T>C non-coding 1 1
152 T>C non-coding 1 1
195 T>C non-coding 1 0
573 ins +1/2 bp non-coding 1 0
4232 T>C (Ile309Thr) ND1 1 0
4703 T>C ND2 1 0
4722 A>G (Thr85Ala) ND2 1 0
4727 A>G ND2 3 1
4745 A>G ND2 1 2
8265 T>C (Leu27Pro) CO2 0 1
9758 T>C CO3 0 1
11404 A>G ND4 1 0
14653 C>T 1 ND6 1 0
14569 G>A ND6 1 0
14766 T>C (Ile7Thr) CYB 2 2
15257 G>A (Asp171Asn) CYB 0 1 1 Novel variation
The 19 different heteroplasmic variations detected are listed in Table 7. A
heteroplasmic mtDNA variation was detected in 13 RM women (27%) and 9 control
women (19%). Of the detected 19 variations, 18 have previously been reported as
polymorphic variations in the Mitomap or mtDB database, while the variation 14653C>T,
a synonymous change in the ND6 gene, has not been previously reported. Altogether, 12
different variations were detected within seven mitochondrial protein coding genes. No
61
heteroplasmic variations were detected within the mitochondrial tRNA and mitochondrial
rRNA regions. Within the non-coding control region seven heteroplasmic variations were
detected. Seven of the variations located within the coding regions are synonymous
changes, while five of the changes are non-synonymous, predicted to change the amino
acid. A non-synonymous change was detected in four RM women and four control
women. The heteroplasmy levels of the non-synonymous variations were studied using
solid-phase minisequencing and the heteroplasmy levels are presented in Table 8.
To study if the 7 variations detected in the mtDNA control region affected the mtDNA
copy number, the mtDNA levels in samples with a variation in the control region (n=8)
were analyzed using quantitative real-time PCR (Q-PCR) and compared to the mtDNA
levels in samples with no heteroplasmic mtDNA variations (n=6). We did not detect
significant differences (p=0.3568) in the mtDNA copy number between samples with a
control region variation and control samples with no variations (Figure 7).
Table 8. Heteroplasmy levels of mtDNA variations.
Level of
Variation Sample heteroplasmy
195 T>C RM patient 85 %
placenta 50 %
4232 T>C (Ile>Thr) RM patient 10 %
4722 A>G (Thr>Ala) RM patient 30 %
8265 T>C (Leu>Pro) control 75 %
11404 A>G RM patient 10 %
placenta <5 %
placenta 15 %
placenta 15 %
14653 C>T RM patient 85 %
placenta 65 %
placenta >95%
placenta 70 %
placenta 40 %
14766 T>C (Ile>Thr) RM patient <5 %
RM patient <5 %
control <95 %
control <5 %
15257 G>A (Asp>Asn) control 25 %
62
2.4.2 Analysis of placental samples
Analysis of human pedigrees segregating pathogenic mtDNA point mutations show large
differences in the proportion of mutant mtDNAs among siblings and between generations
as a heteroplasmic variation is transmitted from the mother to the offspring (Ghosh et al.,
1996; Blok et al., 1997). The rapid changes in mitochondrial DNA allele frequency
observed between generations have been explained by an mtDNA bottleneck in the germ
line. Because of this bottleneck, the mtDNA genotype in the embryo is largely determined
by the genotype of the few mitochondria in the primordial germ cell (Jenuth et al., 1996;
Shoubridge, 2000). Due to the mitochondrial genetic bottleneck new mtDNA mutations
are either lost during transmission or quickly reach very high levels.
Figure 7. MtDNA copy number analysis by Q-PCR in cases with mtDNA control region variations. Eight
lymphocyte samples and one placental sample with at least one variation in the mtDNA control region were
compared to control lymphocyte (n=6) and placental samples (n=6) without mtDNA variations.
To study the transmission of the mtDNA variations from the mother to the offspring
we studied the presence of the previously mentioned variations in placental samples.
Unfortunately, placental samples were available from only three women with a
heteroplasmic mtDNA variation; 195T>C (1 placental sample), 11404A>G (3 placental
samples), and 14653C>T (4 placental samples). By analyzing these 8 placental samples
using dHPLC, we were able to show that the variations were present in the placental
tissue. The heteroplasmy level of these variations was studied using solid-phase
minisequencing (Table 8). In three cases the proportion of the mtDNA variation was
increased in the placenta compared to the mother, and in the remaining cases the
placenta lymphocyte lymphocyte
control
placenta
control
MtDNA copy
number (%)
63
proportion was decreased in the placenta. This result is similar to previous studies
showing variable levels of heteroplasmic variations in the offspring (Ghosh et al., 1996;
Blok et al., 1997).
In addition, the mtDNA level of one placental sample with a control region variation
was compared to the mtDNA levels in placental samples without a control region variation
(n=6) using Q-PCR (Figure 7). No significant differences in the mtDNA copy number
levels were detected.
3 The impact of polymorphisms on phenotype
Single nucleotide polymorphisms (SNPs) are the most common type of human genetic
variation and to date over 3 million SNPs have been reported (Balakrishnan et al., 2005).
Functional SNPs can be categorized as coding SNPs that change amino acids, and
regulatory SNPs that modulate expression or splicing of the gene. However, the role of a
specific polymorphism in a disease or a phenotype may be difficult to assess, especially if
the polymorphism is located in the non-coding regions of the genome. Most
polymorphisms are neutral but a number of SNPs in both coding and non-coding regions
of the genome have also been reported to have phenotypic effects (Suh and Vijg, 2005;
Balakrishnan et al., 2005).
3.1 Variations in the nuclear genome (I, II, III, unpublished)
Most of the variations detected in the candidate genes AMN, TM, EPCR, and p53 were
more frequent than 1% in the control population, which is one of the definitions of a
genetic polymorphism. In recessive disorders it is, however, not uncommon to have carrier
frequencies of known deleterious mutations >1%, particularly in certain populations.
Considering the frequency of miscarriage in the general population, one would expect a
fairly high carrier frequency, if there was a relatively common mutation/variation
underlying miscarriages. Therefore, the frequency of a possible mutation alone is not
always a good indicator of whether a variation is a polymorphism or a mutation. It is also
difficult to determine the function of a SNP on the basis of the nucleotide sequence alone.
This is particularly true when the SNP neither alters an amino acid nor disrupts a well-
characterized motif that affects protein function or structure (Strahan and Read, 2004).
64
In general, nucleotide substitutions within exons may influence the function of the
protein through changes in amino acid composition. Exonic variations that do not result in
a change to the amino acid sequence are called silent mutations, and their phenotypic
effect is often unclear. However, variations in the non-coding regions may also disturb or
alter protein function. SNPs in the promoter region of a gene may influence the
transcriptional activity of the gene by modulating transcription factor binding and
ultimately levels of the gene product. Polymorphisms in the noncoding sequences of
introns and the 3' and 5' UTRs may also result in defects in mRNA processing or stability
(Rebbeck et al., 2004; Malcolm, 2005; Suh and Vijg, 2005). Recently, a growing number
of polymorphisms within the non-coding regions of genes have been shown to alter the
expression or the splicing of a protein (Varley et al., 1998; Attanasio et al., 2001; Wang
XB et al., 2002; Wan et al., 2005; Bourdon et al., 2005). Therefore, some of the
polymorphisms detected in this study may disturb the normal functions of the proteins
encoded by the genes and thereby contribute to the etiology of miscarriage. It would be
interesting to further study the role of some of the variations in RM, especially the
deletion in TM, the duplication in EPCR, and the novel variations c.363G>A and
c.843+11C>T in AMN shown to be more frequent in RM women.
Although the majority of p53 mutations are missense mutations, generally causing a
loss in the ability to bind DNA, intronic variations may also alter the function of the
protein, or the binding sites for interacting proteins (Cho et al., 1994). Single base pair
substitutions in intron 4 of the p53 gene have been shown to disturb the binding of
unidentified transcription factors, resulting in decreased expression of mouse p53
(Beenken et al., 1991). This indicates that the p53 11992 variation may also affect protein
function. In humans, functional analysis has demonstrated elevated functional activity of
variation G13964C in intron 6, which further indicates the importance of intronic
sequences in regulating p53 function (Lehman et al., 2000). In addition, other intronic
base substitutions and duplications/deletions in the p53 gene sequence have been
associated with an increased risk for cancer, indicating that these variations may alter
splicing, expression levels or binding sites, and thereby the function of the protein
(Voglino et al., 1997; Wang-Gohrke et al., 1999; Lacerada et al., 2005). The mechanism
by which the introns regulate either transcription or mRNA stability is, however, unknown
but these results suggest that also the intronic C11992A polymorphism, which we detected
more frequently in patients than controls, may affect the expression level of p53.
65
Alternatively, the variation may affect the protein functions required for protecting the
embryo from teratogens and environmental stress. Further studies are, however, necessary
to the define whether the intronic polymorphism has functional consequences and to
determine if the association with a higher risk for miscarriage is clinically significant.
3.1.1 Possible splice site variations
Approximately 60% of the listed disease causing mutations are missense or nonsense
mutations. In addition, mutations at splice sites make a significant contribution to human
genetic diseases. Approximately 15% of disease causing mutations affect pre-mRNA
splicing through exon skipping, activation of cryptic splice sites, creation of pseudo exons
within introns, or intron retention (Krawczak et al., 1992; Nakai and Sakamoto, 1994;
Baralle and Baralle, 2008). Therefore, we analysed the variations which were not
previously defined as common polymorphisms and the novel sequence variations with the
ESEfinder to see if any of these variations may affect splicing by altering the binding sites
of Ser/Arg-rich (SR) proteins. The program predicted that three single nucleotide
substitutions detected in the AMN gene and the deletion detected in the TM gene may
affect splicing by deleting at least one binding site for known SR-proteins and may
therefore result in exon skipping or intron retention (Table 9 and Figure 8). These
variations may affect splicing of the gene and thereby its function. Conformation of the
phenotypic effect of these variations, however, requires further studies.
The ESEfinder was mainly developed for analysing exonic sequences, but has also
been used to analyze and identify mutations in introns that affect splicing (Gabut et al.,
2005; Stogmann et al., 2006). However, the program only makes predictions about the
effect a variation may have on mRNA splicing. Thus, the presence of a high score motif in
a sequence does not necessarily mean that the sequence determines a splicing enhancer. In
addition, the sequences with the maximum scores are not necessarily the most effective
splicing enhancers. The program currently searches for ESE motifs corresponding to four
SR-proteins (SF2/ASF, SC35, SRp40, and SRp55) but there are several other SR-proteins
for which the ESE motifs have not yet been identified (Cartegni et al., 2003).
66
Table 9. Threshold (Thr) values defined by the ESEfinder and predictive values obtained for normal and
variant alleles for sequence variations predicted to affect the binding sites of SR-proteins.
Figure 8. Diagram of the predicted values obtained by the ESEfinder for the exonic variation AMN
c.829A>G. Graphics of the normal (A) and variant (B) sequence show that the binding site for SF2/ASF
(black bar) present in the normal sequence (GACACCT) is abolished in the variant sequence (GACGCCT).
3.1.2 Couple analysis
The 40 couples included in this study were analysed to determine if some of the possible
functional variations could be detected in both partners of a couple, thereby enabling a
homozygous foetus. In two of the couples both partners had a different exonic or putative
splice site affecting variation in the AMN gene. One couple had AMN variations c.363G>A
and c.1170-6C>T, and another couple had AMN variations c.843+11C>T and
c.1339_1344dup. In these couples there is a 25% chance for every conceptus to be a
compound heterozygote for potentially deleterious AMN mutations in the two alleles.
Additionally, the newly identified 1728+23_+40 deletion in TM was detected in a
heterozygous state in both partners of one couple, enabling homozygosity of the deletion
SF2/ASF SC35 SRp55
Thr =1.956 Thr =2.383 Thr =2.267
AMN c.829A>G
normal allele 2.080 (GACACCT)
variant allele <1.956 (GACGCCT)
AMN c.843+11C>T
normal allele 2.506 (CGCGCGG) 2.584 (GGGCCGCG)
variant allele <1.956 (TGCGCGG) <2.383 (GGGCTGCG)
AMN c.1170-6C>T
normal allele 2.633 (GCCCTCAG) 2.922 (TACGCC)
variant allele <2.383 (GCTCTCAG) <2.267 (TACGCT)
TM c.1728+23_+40del
normal allele 2.742 (GGAGCCTG) 3.831 (TCCGTC)
variant allele sequence deleted sequence deleted
A) B)
67
in the conceptus. This variation, located 23 bp downstream of the exon-intron boundary,
may affect splicing, or alternatively may have an effect on the stability of the mRNA.
Unfortunately, there were no tissue samples from miscarriages experienced by these
couples to study whether these variations were inherited in a homozygous form.
3.2 Variations in the mitochondrial genome (IV)
Although mtDNA is highly polymorphic and both pathogenic and silent variations are
generated in the mtDNA genome, the occurrance of heteroplasmic mtDNA sequence
variations is rare (Jenuth et al., 1996; McKenzie et al., 2004). For example, MELAS and
MERRF mutations are invariably heteroplasmic and homoplasmic mutations in these
genes are expected to be lethal (Howell et al., 2000). As the presence of two different
mtDNA populations in an individual is reported to be rare (Jenuth et al., 1996; DiMauro
and Schon, 2001; McKenzie et al., 2004), the percentage of samples with a heteroplasmic
variation was unexpectedly high in our study. According to our study every fifth control
woman carries a heteroplasmic variation, indicating that heteroplasmy is more common
than previously reported. However, it may be that the high level of homoplasmy is over-
emphasized as most studies rely on techniques that cannot detect low-level heteroplasmic
mtDNA variations (Grzybowski et al., 2000). The fact that we detected heteroplasmy in so
many of the studied samples is most likely due to the sensitivity of the detection method
used in our study. In 8/19 samples (42%) in which the level of heteroplasmy was tested
the ratio of variant sequences was ≤10% or ≥90%, and could not be detected by direct
sequencing.
In this study, a heteroplasmic variation, predicted to change an amino acid, was
detected in four RM women. All of these non-synonymous variations are previously
reported as homoplasmic polymorphisms, and are therefore likely to be non-pathogenic. It
is, however, possible that these variations are neutral when present in a small percentage
of cells, but pathogenic when a threshold is reached or in a homoplasmic state. In general,
the proportion of mutant mtDNAs in a heteroplasmic individual is expected to be very
high before a biochemical or clinical phenotype is observed (Jenuth et al., 1996). Non-
synonymous changes in mitochondrial genes may affect ATP production or apoptosis,
thereby leading to complications during early development. In addition, the mitochondria
are involved in the placental progesterone synthesis required for the maintenance of
68
pregnancy (Tuckey, 2005). One of the detected coding region variations is novel, not
previously reported in the MitoMap or the mtDB database. This variation, 14653C>T in
the ND6 gene is, however, unlikely to be pathogenic as it does not change an amino acid.
In addition, this variation was detected at a heteroplasmy level of 85% in one otherwise
healthy RM woman. Consequently, this novel variation is likely to be neutral.
Analysis of the mtDNA control region revealed 7 heteroplasmic mutations among RM
patients. These variations in the most polymorphic region of the mitochondrial genome
have previously been reported to be polymorphisms. However, mutations in the control
region may affect replication of the mitochondrial genome and such mutations are
suggested to be an important factor in the outcome of a pregnancy. As mitochondrial
replication begins after implantation, replication defects are not likely to compromise
preimplantation embryogenesis, but depending upon mutant load, may lead to post-
implantation
embryonic death (Van Blerkom, 2004). It has been reported that the
localization of mitochondria in the early embryo is strictly regulated and the mitochondrial
distribution is thought to play a role in defining embryonic patterning and long term
viability of the blastomere. Therefore, mtDNA copy number may be an important
determinant of oocyte quality as too few copies would result in maldistribution in the early
embryo (Van Blerkom et al., 2000; Dumollard et al., 2007). To study if the detected
control region variations is affecting mtDNA replication, we analyzed the mtDNA copy
number in these samples. No significant differences in the mtDNA copy number were
shown.
4 Analysis of sex chromosome characteristics (V)
Recent publications have reported associations between RM and skewed XCI in females
and Y chromosome microdeletions in males. Therefore, both of these sex chromosome
characteristics were studied in our patients.
4.1 X chromosome inactivation analysis
In recent years several studies have shown that skewed XCI, defined as preferential
inactivation of one of the two X chromosomes, is increased in women with RM (Lanasa et
al., 1999; Lanasa et al., 2001; Uehara et al., 2001; Kuo et al., 2008). Normally inactivation
69
of the X chromosomes occurs randomly on both the maternal and paternal X
chromosomes during the late blastocyst state of embryogenesis (Lyon, 1961). The
suggested explanation for the association between skewed XCI and RM is an unknown X
chromosomal mutation resulting in both preferential inactivation of the mutated X and
lethality in male embryos carrying the mutation (Lanasa et al., 1999; Uehara et al., 2001;
Bagislar et al., 2006).
To study the role of XCI in RM in our Finnish patients the XCI patterns were
examined in 46 women with RM and 95 control women. The XCI ratio was determined by
analysing the methylation status of the X-linked AR gene using a methylation sensitive
restriction enzyme that digests only nonmethylated DNA segments (Figure 9). Samples
homozygous at the AR (CAG)n locus were excluded from the X-inactivation study due to
the inability to distinguish between the two alleles. 3/46 women with RM and 14/95
control women were homozygous for the AR (CAG)n polymorphism and excluded from
further analysis. The ratio of inactivation between the two X chromosomes was calculated
for the remaining 43 female patients and 81 controls.
Figure 9. Fluorescence based X-inactivation analysis using a PCR based AR methylation assay. The
methylation sensitive restriction enzyme HpaII digests and prevents amplification of nonmethylated (active
X) DNA segments. (a) An example of equivalent inactivation of both alleles. (b) An example of extremely
skewed inactivation of the alleles. Undigested DNA is used as control for peak intensities of the alleles.
There are several possible reasons for the discrepancy between this study and those
previously reported. Firstly, it is possible that the number of patients is too small to
provide statistically reliable results if skewed XCI is the cause of RM in only a small
subset of patients. However, previous studies have included approximately the same
number of subjects and have been able to show an association between skewed XCI and
undigested undigested
Hpa II digested Hpa II digested
a) b)
70
RM (Lanasa et al., 1999; Uehara et al., 2001). Secondly, previous studies have used
different inclusion criteria, which may affect the results. Our study included
patients with
three or more pregnancy losses. In contrast, other studies have included patients with only
two pregnancy losses (Lanasa et al., 1999; Lanasa et al., 2001; Kim et al., 2004). Finally,
in our study the patients and controls were of same age, the mean age in both groups being
31 years, whereas in other studies the patients were significantly older than their
controls
(Lanasa et al., 1999, Lanasa et al., 2001). Because skewed XCI increases in peripheral
lymphocytes with age due to depletion of the haematopoietic stem cell pool the age
discrepancy in other studies could have resulted in a bias towards an increase in skewed
XCI in RM cases when compared with younger controls (Sharp et al., 2000; Hogge et al.,
2007).
There are, however, also studies that have not detected an association between skewed
XCI and RM (Sullivan et al., 2003; Kim et al., 2004; Hogge et al., 2007; Dasoula et al.,
2008). Beever and coworkers (2003) evaluated the XCI status in over 200 women with
RM and did not detect an association between skewed XCI and RM. Futhermore, they
detected a significant excess of boys in live births among women with RM, which is
contrary to expectations if the skewed XCI is due to X-linked lethal mutations (Beever et
al., 2003). In some women skewed XCI may be due to the limited number of precursor
cells present during the late blastocyst stage when the X chromosome inactivation process
begins. At this developmental stage, there are approximately only 10–20 cells that will
subsequently form the embryo; the other cells differentiate into extra embryonic tissues.
Thus, some women may show skewed XCI due to chance alone (Lyon, 1971; Belmont,
1996; Gale et al., 1997). Alternatively, it is possible that skewed XCI is a heritable genetic
trait that is detected in family members in multiple generations. Genetic loci that influence
the selection of the inactivated X chromosome have been detected in mice (Percec et al.,
2002; Chadwick et al., 2006). Whether similar loci exist in humans
is not clear, but skewed
XCI has been mapped in humans to Xq28 (Pegoraro et al., 1997; Naumova et al., 1998). In
addition, skewed XCI appears to be inherited in some families and in these families
skewed XCI may be caused by a genetic variation that is not related to RM (Naumova et
al., 1996).
71
4.2 Y chromosome microdeletion analysis
To date the knowledge of male contribution to RM is largely limited to karyotyping.
Karyotyping gives an overview of the whole genome, but reveals only large chromosomal
rearrangements. Submicroscopic duplications and deletions, such as Yq microdeletions,
remain unidentified using this method. Yq microdeletions are reported to cause
spermatogenetic failure and male infertility (Vogt et al., 1996; Hopps et al., 2003). There
are at least three regions, AZFa, AZFb, and AZFc, on the long arm of the Y chromosome,
in which deletions are associated with varying degree of spermatogenic failure. Although
it is widely accepted that deletions of these regions contribute to male infertility, the
function of the genes located in the AZF regions is poorly understood (Krausz and
McElreavey, 1999; Noordam and Repping, 2006; Ferlin et al., 2007). It has been reported
that men with Y chromosome microdeletions have low sperm counts and a higher
incidence of sperm chromosomal abnormalities (Foresta et al., 2005). Because sperm
chromosome abnormalities have been found in some men of RM couples (Rubio et al,
1999; Al-Hassan et al., 2005) Dewan and coworkers (2006) studied if Y chromosome
microdeletions are associated with RM. They reported that a significant portion of male
partners (82%) of women with RM had Y chromosomal microdeletions, and suggest that
the Y chromosomal genes may be important for early embryonic development in male
embryos (Dewan et al., 2006).
In our study we investigated if a similar association between Y chromosome
microdeletions and RM was present in our set of Finnish patients. We tested 40 male
partners of women with unexplained RM for the presence of Y chromosome
microdeletions. In addition, to the four STSs used in the previous study (Dewan et al.,
2006) we tested our patients using 33 additional STSs to enable the analysis of all AZF
regions. One STS was used to test for the presence of the short arm of the Y chromosome,
the other 32 STS loci were located on the long arm of the Y chromosome. Amplification
products for all of the 37 STSs used were detected in all tested patients, and thereby we
did not detect any Y chromosome microdeletions in our samples.
Although our study was performed on a bigger set of samples compared to the study
by Dewan et al. (2006), our study revealed no microdeletions and therefore does not
support the previously reported findings. It is possible that the involvement of
microdeletions in the aetiology of RM differs between populations and this may be the
72
cause of different results between the studies. It is known that different populations have
different Y chromosome haplogroups and it has been shown that the susceptibility to Y
chromosomal microdeletions varies between different haplogroups (Arredi et al., 2007;
Yang et al., 2008). There may also be some technical problems affecting the results of the
Dewan et al. (2006) study (Noordam et al., 2006). These technical factors could be the
underlying cause of the discrepancy between the previously reported results and the results
of our study. According to the NCBI UniSTS Database, the STS marker DYS262 is
located on the short arm of the Y chromosome, and the markers DYS220, DYF86S1, and
DYF85S1 on the long arm in the AZFc region. Dewan et al. (2006) report that they found
patients with both DYS262 and an additional marker deleted. This indicates that these
patients would have multiple microdeletions or alternatively a region as big as 15 Mb
deleted. This is, however, unlikely in otherwise healthy males.
According to guidelines provided by the European Academy of Andrology (EAA) and
the European Molecular Genetics Quality Network (EMQN) for diagnostic laboratories,
the use of 6 STSs should be able to detect up to 95% of all reported Y microdeletions in
the AZF regions (Simoni et al., 2004). We tested our male patients using all the STSs
recommended in the guidelines, and in addition, over 30 other markers spanning the Y
chromosome. Therefore, it is unlikely that we would have missed at least any of the
known deletions.
Although we did not detect any microdeletions, it is important to keep in mind that a
normal lymphocyte karyotype does not exclude sperm abnormalities. High rates of
chromosomal abnormalities in sperm have been observed in RM patients with normal
lymphocyte karyotypes because the highly repetitive structure of the Y chromosome
sequence predisposes to de novo deletions (Rubio et al, 1999; Al-Hassan et al., 2005).
This study was performed using DNA extracted from peripheral blood and therefore, it is
possible that abnormalities leading to miscarriage have arisen in the spermatocytes during
spermatogenesis through de novo mutations.
73
5 Limitations of the study
When considering the known causes of miscarriages, the aetiology of RM seems very
heterogeneous. Therefore, it is possible that mutations in the candidate genes of this study
cause miscarriage in only a small subpopulation of couples with miscarriages and one may
argue that more couples would be needed to determine the exact role of the candidate
genes in RM. After excluding the common variations, the mutation rate was low in the
candidate genes. This may be due to the fact that mutations in these genes are rare in
general or that they are a rare reason for RM. Screening more patients and controls may
also have revealed a statistical difference in the frequencies of the variations between
patients and controls. Additionally, a larger number of controls would be needed to
calculate the actual frequency of the novel variations and to determine their significance,
as when studying recessive conditions or complex diseases some of the controls are also
expected to be carriers of the disease causing mutation or susceptibility alleles.
Additionally, we also may have missed some mutations in the candidate genes.
Although in theory dHPLC is a highly sensitive and specific method for mutation
detection, the sensitivity of mutation detection by heteroduplex analysis is unlikely to be
100%. There are some rare instances where mutations may not be detected. Some
variations can only be detected at one unique temperature, and variations located within a
GC-rich sequence of a fragment with otherwise normal nucleotide ratios may in some
cases not be detected due to the high melt temperature of GC-rich pockets (Xiao and
Oefner, 2001). In this study one variation in TM (c.1418C>T) was initially not detected
using dHPLC but by sequencing a random selection of samples. Another potential
problem relates to the detectability of heteroplasmic mtDNA mutations in samples with a
proportion of <5% or >95% mutant mtDNA, because heteroduplex analysis becomes
unreliable beyond those limits (Biggin et al., 2004; Lim K et al., 2008).
Another limitation of this study is the use of blood samples for detection of
heteroplasmic mtDNA variations. It has been shown that the degree of heteroplasmy for
some mitochondrial DNA mutations varies considerably among tissues. Studies have
reported a decrease in the proportion of mutant mtDNA in blood by approximately 0.5%
to 2% per year (Howell et al., 2000; Rajasimha et al., 2008; Lim et al., 2008). As a
consequence blood may contain lower levels of the mutation compared to other tissues,
making the detection of low-level mutations demanding.
74
Conclusions
The aim of this study was to find new genetic causes for RM, a pregnancy complication
that affects around 1% of couples trying to conceive. To date there are many factors
known to cause RM, but the underlying cause of a couple´s miscarriages often remains
unresolved as RM is a very heterogeneous condition. The identification of the underlying
causes of miscarriage is important for developing more successful treatments for women
experiencing RM and thereby preventing subsequent miscarriages. Identification of
underlying causes also helps to identify pregnant women that are at a risk of having a
miscarriage. This study aimed to identify new genetic factors underlying RM by studying
candidate genes - either identified from animal models or regulating functions important in
different stages of early foetal developmental - and sex chromosome characteristics
previously reported to be associated with RM.
By screening the couples for mutations in the AMN, TM and EPCR genes, two novel
AMN variations were shown to be associated with RM. One variation, c.363G>A, is an
exonic synonymous variation, and the other, c.843+11C>T is an intronic variation
predicted to affect splicing. These variations may increase the risk of RM in carrier
women. Further studies would, however, be required to confirm the phenotypic effect of
these variations. In addition, four interesting exonic or potential splice site disrupting
variations were detected in these genes; c.1339_1344dup and c.1170-6C>T in AMN,
c.1728+23_+40del in TM, and c.323-9_336dup in EPCR. In one couple both partners were
heterozygous for a deletion in the TM gene and in two couples both partners had different
exonic or potential splice site disrupting AMN variations. These variations may play a role
in the miscarriages of these couples. However, the phenotypic effects of the potentially
deleterious variations cannot be determined without further investigations. It would be
interesting to study these variatons in a larger sample set or alternatively, in
foetal/placental samples. However, most of the variations detected in these genes are
likely to be silent polymorphisms and, taken together, the results of this study suggest that
mutations in AMN, TM, and EPCR do not have a major role in RM in the patients studied,
even though their contribution to RM cannot be totally excluded.
75
Additionally, a polymorphism in the p53 gene, C11992A, was shown to be associated
with RM in Finnish patients. The results indicate that women with a C/A or A/A genotype
have a twofold higher risk for RM than women with a C/C genotype. As the
polymorphism was, however, detected in a large number of controls as well it is likely to
be a contributing factor increasing the risk of RM in combination with other sequence
variations or environmental factors, rather than the sole cause of RM. It is also possible
that, instead of the 11992 locus, another genetic variation linked to this locus, is the actual
susceptibility locus increasing the risk for RM. While the exact role of p53 in development
is unclear, it is difficult to predict whether the variation disturbs functions needed for
normal embryonic development. The results of the p53 expression analysis indicate that
the miscarriages experienced by the studied couples are not due to altered p53 levels in the
placenta.
When screening the mitochondrial genome heteroplasmic mtDNA variations were
detected in 13 RM women, which was unexpected as heteroplasmic variations are reported
to be rare. Most of these variations are previously reported as polymorphisms and the
novel variation is likely to be non-pathogenic as it is synonymous. However, the role of
these variations in the aetiology of miscarriage cannot be excluded as some variations may
be pathogenic only when a certain threshold is reached. The transmission of the variations
was studied by analysing placental samples and the result is similar to previous studies
showing that a heteroplasmic variation in a mother is transmitted to her offspring in
different proportions. This supports the hypothesis that a woman with no disease
phenotype can carry a low-level heteroplasmic variation that may cause foetal death when
transmitted to the foetus in a higher proportion. However, the variations detected in this
study were judged to be polymorphisms, and not likely to contribute to the miscarriages
experienced by the women studied. This, however, does not exclude the role of other
heteroplasmic mtDNA variations in miscarriage.
In this study sex chromosome abnormalities, recently reported to be associated with
RM, were also investigated to study if these factors could underlie the RM in Finnish
patients. The female patients were studied for skewed X chromosome inactivation and the
male partners were studied for Y chromosome microdeletions. Based on the results of the
study it was concluded that there is no association between skewed X chromosome
inactivation or Y chromosome microdeletions and RM in Finnish patients.
76
Although no major genetic cause explaining the miscarriages in the patients could be
identified, the results of the genealogic studies indicate that a subset of patients possibly
have a common genetic cause underlying their miscarriages. The results of the genealogic
studies suggest that there may be an eastern enrichment of ancestral birthplaces in the
province of Kuopio, which was unexpected as the patients were collected from Helsinki. It
would be interesting to collect more patients from this region for further genetic studies.
77
Acknowledgements
This study was carried out at the Folkhälsan Institute of Genetics, and the Department of
Medical Genetics, University of Helsinki during 2004-2009. I wish to thank the former
and current heads of these institutes, Professors Anna-Elina Lehesjoki, Kristiina Aittomäki
and Päivi Peltomäki for providing excellent working facilities in Biomedicum Helsinki.
I am grateful to the following organizations for financially supporting my work: the
Folkhälsan Research Foundation, the Sigrid Juselius Foundation, the Biomedicum
Helsinki Foundation, the Emil Aaltonen Foundation, the Maud Kuistila Memorial
Foudation, the University of Helsinki Funds, the Oskar Öflund Foundation, and Svenska
Kulturfonden.
I wish to express my gratitude to my supervisor docent Kristiina Aittomäki for giving me
the opportunity to work in her research group and letting me participate in a very
interesting project. I am grateful to Kristiina for giving me responsibility for my projects
and for her confidence in me throughout this project. I guess you knew I would end up
defending my thesis eventhough I wasn´t so convinced myself all the time. I would also
like to thank Kristiina for supporting me in my aim of becoming a medical geneticist.
Docent Leena Anttila and docent Nina Horelli-Kuitunen, the official reviewers of this
thesis work, are warmly thanked for critical review and for valuable suggestions that
improved this thesis.
Professor Päivi Peltomäki is warmly thanked for accepting the important post of custos at
the dissertation.
I do owe enormous thanks to Jodie Painter, my supervisor in the lab, for encouragement
and for always being willing to help me. Jodie is thanked for help in the lab, especially in
the beginning of the project, invaluable help during a number of writing processes and
teaching me a scientific way of thinking. Jodie is also thanked for revising the language of
this thesis work and helpful comments to improve it.
I am deeply grateful to all the co-authors and collaborators in this study for their valuable
contributions to this thesis work. Veli-Matti Ulander and Risto Kaaja are thanked for
collecting and treating the patients included in this study, for clinical expertise and for a
fruitful and nice collaboration. Ralf Bützow is thanked for providing his expertise in
immunohistochemical studies. Anu Wartiovaara (née Suomalainen) and Alexandra Götz
are thanked for collaboration within the mitochondrial field and valuable contributions to
my thesis work under a tight schedule. Sarah Ariansen is warmly thanked for valuable
collaboration and for welcoming me to visit their lab in Oslo. Taru Ahvenainen is thanked
for careful lab work with the TM gene.
I am thankful to research nurses Katja Kuosa and Eija Kortelainen for providing me with
exact information about the study subjects. Aila Riikonen, Jaana Welin-Haapamäki,
Stephan Keskinen, Marjatta Valkama, and Solveig Halonen are acknowledged for helping
with practical matters during the years.
78
I would like to thank Per-Henrik Groop for encouragement and interest in my studies,
thesis project, and future plans. Hannele Laivuori is thanked for fruitful collaboration
within the field of reproductive medicine and a positive attitude towards my work.
During my time working in the lab and writing my thesis I have greatly enjoyed the
company of my research colleagues at Folkhälsan Institute of Genetics. Maria, my fellow
PhD student in our small group, your company and support have been really important for
me, thank you. Vilma is thanked for friendship, interesting converations and for sharing
the frustration of working with the WAVE. I would like to thank Anne S, Eija, and Nina
for friendship and helping me with all kinds of questions, especially when finishing this
thesis project. Kati and Ann-Liz are thanked for their positive attitude towards most things
in life, and for making me smile :) Peter and Carol, the FHRM team, thank you for
amusing planning meetings during the past years. Saara, Mervi, Jaakko, Otto, Maria K,
Hanna V, Elina, Reetta, and all the former and present members of the Folkhälsan team
are thanked for useful advice, friendship, and for creating a nice and unique working
atmosphere.
I also wish to thank Joni, Annu, Helena, Pia, and Jeanette, with whom I have been sharing
the office room, for nice conversations about traveling and renovations, and for patience in
listening to and sharing my frustration when things were not going the way I wished. It
became quite empty in our room since you left.
I want to express my warmest thanks to my dear friends, Minttu, Ammi, and Stina for
always being there when I have needed you. Minja, Jenny, Anna, Mariann, Emma, and
Julia are warmly thanked for friendship, for scientific and non-scientific discussions
during both lunch hours and spare-time, and for making studing biology fun! It has been a
great pleasure to get to know you. I wish to sincerely thank Juhis and Toffe for friendship
and for convincing me since ages ago that I should do a PhD.
My deepest appreciation goes to my family for supporting me in all my decisions and an
encouraging attitude towards my studies. I own my warmest thanks to Miro, my friend,
my love, for his love and support during these years and for taking my thoughts off work.
Thank you for taking care of the penguin.
Finally, all the patients who have voluntarily participated in this study, thereby making it
possible, are warmly thanked.
Helsinki, February 2009
79
References
Aarli A, Kristoffersen EK, Jensen TS, Ulvestad E, Matre R (1997). Suppressive effect on lymphoproliferation in vitro by soluble annexin II released from isolated placental membranes. Am J Reprod Immunol. 38:313-319.
Abdalla HI, Burton G, Kirkland A, Johnson MR, Leonard T, Brooks AA, Studd JW (1993). Age, pregnancy amd miscarriage: uterine versus ovarian factors. Hum Reprod. 8:1512-1517.
Aittomäki K, Lucena JL, Pakarinen P, Sistonen P, Tapanainen J, Gromoll J, Kaskikari R, Sankila EM, Lehväslaiho H, Engel AR, Nieschlag E, Huhtaniemi I, de la Chapelle A (1995). Mutation in the follicle-stimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell. 82:959-968.
Aittomäki K, Herva R, Stenman UH, Juntunen K, Ylöstalo P, Hovatta O, de la Chapelle A (1996). Clinical features of primary ovarian failure caused by a point mutation in the follicle-stimulating hormone receptor gene. J Clin Endocrinol Metabol. 81:3722-3726.
Aldrich CL, Stephenson MD, Karrison T, Odem RR, Branch DW, Scott JR, Schreiber JR, Ober C (2001). HLA-G genotypes and pregnancy outcome in couples with unexplained recurrent miscarriage. Mol Hum Reprod. 7:1167-1172.
Al-Hassan S, Hellani A, Al-Shahrani A, Al-Deery M, Jaroudi K, Coskun S (2005). Sperm chromosomal abnormalities in patients with unexplained recurrent abortions. Arch Androl. 51:69-76.
Aminoff M, Carter JE, Chadwick RB, Johnson C, Gräsbeck R, Abdelaal MA, Broch H, Jenner LB, Verroust PJ, Moestrup SK, de la Chapelle A, Krahe R (1999). Mutations in CUBN, encoding the intrinsic factor-vitamin B12 receptor, cubilin, cause hereditary megaloblastic anaemia 1. Nat Genet. 21:309-313.
Antonarakis SE, Beckmann JS (2006). Mendelian disorders deserve more attention. Nat Rev Genet. 7:277-282.
Armstrong JF, Kaufman MH, Harrison DJ, Clarke AR (1995). High-frequency developmental abnormalities in p53-deficient mice. Curr Biol. 5:931-936.
Arredi B, Ferlin A, Speltra E, Bedin C, Zuccarello D, Ganz F, Marchina E, Stuppia L, Krausz C, Foresta C (2007). Y-chromosome haplogroups and susceptibility to azoospermia factor c microdeletion in an Italian population. J Med Genet. 44:205-208.
Arredondo F, Noble LS (2006). Endocrinology of recurrent pregnancy loss. Semin Reprod Med. 24:33-39.
Ashley MV, Laipis PJ, Hauswirth WW (1989). Rapid segregation of heteroplasmic bovine mitochondria. Nucleic Acids Res. 17:7325-7331.
Attanasio C, de Moerloose P, Antonarakis SE, Morris MA, Neerman-Arbez M (2001). Activation of multiple cryptic donor splice sites by the common congenital afibrinogenemia mutation, FGA IVS4 + 1 G->T. Blood. 97:1879-1881.
Azofeifa J, Voit T, Hubner C, Cremer M (1995). X-chromosome methylation in manifesting and healthy carriers of dystrophinopathies: concordance of activation ratios among first degree female relatives and skewed inactivation as cause of the affected phenotypes. Hum Genet. 96:167-176.
Baek KH (2004). Aberrant gene expression associated with recurrent pregnancy loss. Mol Hum Reprod. 10:291-297.
Baek KH, Choi BC, Lee J-H, Choi H-K, Lee S-H, Kim J-W, Hill JA, Chung H-M, Ko JJ, Cha KY (2002) Comparison of gene expression at the feto-maternal interface between normal and recurrent pregnancy loss patients. Reprod Fertil Dev. 14:235–240.
Bagislar S, Ustuner I, Cengiz B, Soylemez F, Akyerli CB, Ceylaner S, Ceylaner G, Acar A, Ozcelik T (2006). Extremely skewed X-chromosome inactivation patterns in women with recurrent spontaneous abortion. Aust N Z J Obstet Gynaecol. 46:384-387.
80
Bajekal N, Li TC (2000). Fibroids, infertility and pregnancy wastage. Hum Reprod Update. 6:614-620.
Balakrishnan VS, Rao M, Jaber BL; DialGene Consortium (2005). Genomic medicine, gene polymorphisms, and human biological diversity. Semin Dial. 18:37-40.
Balint EE, Vousden KH (2001). Activation and activities of the p53 tumour suppressor protein. Br J Cancer. 85:1813-1823.
Ballabio A (1993). The rise and fall of positional cloning? Nat Genet. 3:277-279.
Baralle D, Baralle M (2005). Splicing in action: assessing disease causing sequence changes. J Med Genet. 42:737-748.
Beddington RS, Robertson EJ (1999). Axis development and early asymmetry in mammals. Cell. 96:195-209.
Beenken SW, Karsenty G, Raycroft L, Lozano G (1991). An intron binding protein is required for transformation ability of p53. Nucleic Acids Res. 19:4747-4752.
Beever CL, Stephenson MD, Peñaherrera MS, Jiang RH, Kalousek DK, Hayden M, Field L, Brown CJ, Robinson WP (2003). Skewed X-chromosome inactivation is associated with trisomy in women ascertained on the basis of recurrent spontaneous abortion or chromosomally abnormal pregnancies. Am J Hum Genet. 72:399-407.
Belmont JW (1996). Genetic control of X inactivation and processes leading to X-inactivation skewing. Am J Hum Genet. 58:1101–1108.
Bergstrom CT, Pritchard J (1998). Germline bottlenecks and the evolutionary maintenance of mitochondrial genomes. Genetics. 149:2135-2146.
Biggin A, Henke R, Bennetts B, Thorburn DR, Christodoulou J (2005). Mutation screening of the mitochondrial genome using denaturing high-performance liquid chromatography. Mol Genet Metab. 84:61-74.
Biguzzi E, Merati G, Liaw PC, Bucciarelli P, Oganesyan N, Qu D, Gu JM, Fetiveau R, Esmon CT, Mannucci PM, Faioni EM (2001). A 23bp insertion in the endothelial protein C receptor (EPCR) gene impairs EPCR function. Thromb Haemost. 86:945-948.
Blok RB, Gook DA, Thorburn DR, Dahl HH (1997). Skewed segregation of the mtDNA nt 8993 (T-->G) mutation in human oocytes. Am. J. Hum. Genet. 60:1495-1501.
Bond GL, Hu W, Levine AJ (2005). MDM2 is a central node in the p53 pathway: 12 years and counting. Curr Cancer Drug Targets. 5:3-8.
Bose P, Black S, Kadyrov M, Weissenborn U, Neulen J, Regan L, Huppertz B (2005). Heparin and aspirin attenuate placental apoptosis in vitro: implications for early pregnancy failure. Am J Obstet Gynecol. 192:23-30.
Bourdon JC, Fernandes K, Murray-Zmijewski F, Liu G, Diot A, Xirodimas DP, Saville MK, Lane DP (2005). p53 isoforms can regulate p53 transcriptional activity. Genes Dev. 19:2122-2137.
Brown CJ, Robinson WP (2000). The causes and consequences of random and non-random X chromosome inactivation in humans. Clin Genet. 58:353-363.
Brown SD (1998). Mouse models of genetic disease: new approaches, new paradigms. J Inherit Metab Dis. 21:532-539.
Brown SD, Nolan PM (1998). Mouse mutagenesis-systematic studies of mammalian gene function. Hum Mol Genet. 7:1627-1633.
Bulletti C, Flamigni C, Giacomucci E (1996). Reproductive failure due to spontaneous abortion and recurrent miscarriage. Hum Reprod Update. 2:118-136.
Bussen S, Sutterlin M, Steck T (1999). Endocrine abnormalities during follicular phase in women with recurrent spontaneous abortion. Hum Reprod. 14:18-20.
Carrington B, Sacks G, Regan L (2005). Recurrent miscarriage: pathophysiology and outcome. Curr Opin Obstet Gynecol. 17:591-597.
Cartegni L, Wang J, Zhu Z, Zhang MQ, Krainer AR (2003). ESEfinder: A web resource to identify exonic splicing enhancers. Nuc Acid Research. 31:3568-3571.
81
Carter AM (2007). Animal models of human placentation--a review. Placenta. 28(Suppl):S41-47.
Chadwick LH, Pertz LM, Broman KW, Bartolomei MS, Willard HF (2006). Genetic control of X chromosome inactivation in mice: definition of the Xce candidate interval. Genetics. 173:2103-2110.
Chavatte-Palmer P, Guillomot M (2007). Comparative implantation and placentation. Gynecol Obstet Invest. 64:166-174.
Cheah PL, Looi LM (2001). p53: an overview of over two decades of study. Malays J Pathol. 23:9-16.
Cho Y, Gorina S, Jeffrey PD, Pavletich NP (1994). Crystal structure of a p53 tumor suppressor-DNA complex: understanding tumorigenic mutations. Science. 265:346-355.
Choi H-K, Choi BC, Lee S-H, Kim J-W, Cha KY, Baek K-H (2003). Expression of angiogenesis- and apoptosis-related genes in chorionic villi derived from recurrent pregnancy patients. Mol Reprod. Dev 66:24–31.
Choi J, Donehower LA (1999). p53 in embryonic development: maintaining a fine balance. Cell Mol Life Sci. 55:38-47.
Clark DA, Coulam CB, Daya S, Chaouat G (2001). Unexplained sporadic and recurrent miscarriage in the new millennium: a critical analysis of immune mechanisms and treatments. Hum Reprod Update. 7:501-511.
Clifford K, Rai R, Watson H, Regan L (1994). An informative protocol for the investigation of recurrent miscarriage: preliminary experience of 500 consecutive cases. Hum Reprod. 9:1328-1332.
Cnattingius D, Signorello LB, Anneren G, Clausson B, Ekblom A, Ljunger E, Blot WJ, McLaughlin JK. Petersson G, Rane A, Granath F (2000). Caffein intake and the risk of first-trimester spontaneous abortion. N Eng J Med. 343:1839-1845.
Coan PM, Burton GJ, Ferguson-Smith AC (2005) Imprinted genes in the placenta--a review. Placenta. 26 Suppl A:S10-20.
Collins FS (1992). Positional cloning: let's not call it reverse anymore. Nat Genet. 1:3-6.
Collins FS (1995). Positional cloning moves from perditional to traditional. Nat Genet. 9:347-350.
Collins FS, Patrinos A, Jordan E, Chakravarti A, Gesteland R, Walters L (1998). New goals for the U.S. Human Genome Project: 1998-2003. Science. 282:682-689.
Costeas PA, Koumouli A, Giantsiou-Kyriakou A, Papaloizou A, Koumas L (2004). Th2/Th3 cytokine genotypes are associated with pregnancy loss. Hum Immunol. 65:135-141.
Coulam CB, Kay C, Jeyendran RS (2006). Role of p53 codon 72 polymorphism in recurrent pregnancy loss. Reprod Biomed Online. 12:378-382.
Coulam CB, Jeyendran RS (2008). Vascular endothelial growth factor gene polymorphisms and recurrent pregnancy loss. Am J Reprod Immunol. 59:301-305.
Couto E, Barini R, Zaccaria R, Annicchino-Bizzacchi JM, Passini Junior R, Pereira BG, Silva JC, Pinto e Silva JL (2005) Association of anticardiolipin antibody and C677T in methylenetetrahydrofolate reductase mutation in women with recurrent spontaneous abortions: a new path to thrombophilia? Sao Paulo Med J. 123:15-20.
Crawley JT, Gu JM, Ferrell G and Esmon CT (2002). Distribution of endothelial cell protein C/activated protein C receptor (EPCR) during mouse embryo development. Thromb Haemost. 88:259-266.
Cross JC (2006). Placental function in development and disease. Reprod Fertil Dev. 18:71-76.
Cross JC, Werb Z and Fisher SJ (1994). Implantation and the placenta: key pieces of the development puzzle. Science. 266:1508-1518.
Daher S, Shulzhenko N, Morgun A, Mattar R, Rampim GF, Camano L, DeLima MG (2003). Associations between cytokine gene polymorphisms and recurrent pregnancy loss. J Reprod Immunol. 58:69-77.
82
Dahlbäck B, Villoutreix BO (2005). The anticoagulant protein C pathway. FEBS Lett. 579:3310-3316.
Dalton CF, Laird SM, Estdale SE, Saravelos HG, Li TC (1998). Endometrial protein PP14 and CA-125 in recurrent miscarriage patients; correlation with pregnancy outcome. Hum Reprod. 13:3197–3202.
Dasoula A, Kalantaridou S, Sotiriadis A, Pavlou M, Georgiou I, Paraskevaidis E, Makrigiannakis A, Syrrou M (2008). Skewed X-chromosome inactivation in Greek women with idiopathic recurrent miscarriage. Fetal Diagn Ther. 23:198-203.
den Dunnen JT, White SJ (2006). MLPA and MAPH: sensitive detection of deletions and duplications. Curr Protoc Hum Genet. Chapter 7.14.
Devine DH, Whitman-Elia G, Best RG, Edwards JG (2000). Paternal paracentric inversion of chromosome 2: a possible association with recurrent pregnancy loss and infertility. J Assist Reprod Genet. 17:293–296.
Dewan S, Puscheck EE, Coulam CB, Wilcox AJ, Jeyendran RS (2006). Y-chromosome microdeletions and recurrent pregnancy loss. Fertil Steril. 85:441-445.
Dhont M (2003). Recurrent miscarriage. Curr Womens Health Rep. 3:361-366.
DiMauro S (2001). Lessons from mitochondrial DNA mutations. Semin. Cell Dev. Biol. 12:397-405.
DiMauro S (2007) Mitochondrial DNA medicine. Biosci Rep. 27:5-9.
DiMauro S, Schon EA (2001). Mitochondrial DNA mutations in human disease. Am J Med Genet. 106:18-26.
Dittman WA, Majerus PW (1989). Sequence of a cDNA for mouse thrombomodulin and comparison of the predicted mouse and human amino acid sequences. Nucleic Acids Res 17:802.
Dizon-Townson DS, Lu J, Morgan TK, Ward KJ (2000). Genetic expression by fetal chorionic villi during the first trimester of human gestation. Am J Obstet Gynecol. 183:706-711.
Doggen CJ, Kunz G, Rosendaal FR, Lane DA, Vos HL, Stubbs PJ, Manger Cats V, Ireland H (1998). A mutation in the thrombomodulin gene, 127G to A coding for Ala25Thr, and the risk of myocardial infarction in men. Thromb Haemost. 80:743-748.
Dominguez-Rojas V, de Juanes-Pardo JR, Astasio-Arbiza P, Ortega-Molina P, Gordillo-Florencio E (1994). Spontaneous abortion in a hospital population: are tobacco and coffee intake risk factors? Eur J Epidemiol. 10:665-668.
Dumollard R, Duchen M, Carroll J (2007). The role of mitochondrial function in the oocyte and embryo. Curr Top Dev Biol. 77:21-49.
Dumont P, Leu JI, Della Pietra AC 3rd, George DL, Murphy M (2003). The codon 72 polymorphic variants of p53 have markedly different apoptotic potential. Nat Genet. 33:357-365.
Dvash T, Ben-Yosef D, Eiges R (2006). Human embryonic stem cells as a powerful tool for studying human embryogenesis. Pediatr Res. 60:111-117.
Egozcue S, Blanco J, Vendrell JM, Garcia F, Veiga A, Aran B, Barri PN, Vidal F, Egozcue J (2000). Human male infertility: chromosomal anomalies. Meiotic disorders, abnormal spermatozoa and recurrent abortion. Hum Reprod Update. 6:93-105.
Empson M, Lassere M, Craig JC, Scott JR (2002). Recurrent pregnancy loss with antiphospholipid antibody: a systematic review of therapeutic trials. Obstet Gynecol. 99(1):135-144.
Esmon CT (1989). The roles of protein C and thrombomodulin in the regulation of blood coagulation. J Biol Chem. 264:4743-4746.
Esmon CT (2000). The endothelial cell protein C receptor. Thromb Haemost. 83:639-643.
Faioni EM, Franchi F, Castaman G, Biguzzi E, Rodeghiero F (2002). Mutations in the thrombomodulin gene are rare in patients with severe thrombophilia. Br J Haematol. 118:595-599.
83
Farquharson RG, Quenby S, Greaves M (2002). Antiphospholipid syndrome in pregnancy: a randomized, controlled trial of treatment. Obstet Gynecol. 100(3):408-413.
Fazel A, Vincenot A, Malassine A, Soncin F, Gaussem P, Alsat E, Evain-Brion D (1998). Increase in expression and activity of thrombomodulin in term human syncytiotrophoblast microvilli. Placenta. 19:261-268.
Ferlin A, Arredi B, Speltra E, Cazzadore C, Selice R, Garolla A, Lenzi A, Foresta C (2007). Molecular and clinical characterization of y chromosome microdeletions in infertile men: a 10-year experience in Italy. J Clin Endocrinol Metab. 92:762-770.
Foka ZJ, Lambropoulos AF, Saravelos H, Karas GB, Karavida A, Agorastos T, Zournatzi V, Makris PE, Bontis J, Kotsis A (2000) Factor V leiden and prothrombin G20210A mutations, but not methylenetetrahydrofolate reductase C677T, are associated with recurrent miscarriages. Hum Reprod. 15:458-462.
Foresta C, Garolla A, Bartoloni L, Bettella A, Ferlin A (2005). Genetic abnormalities among severely oligospermic men who are candidates for intracytoplasmic sperm injection. J Clin Endocrinol Metab. 90:152-156.
Franco B, Ballabio A (2006) X-inactivation and human disease: X-linked dominant male-lethal disorders. Curr Opin Genet Dev. 16:254-259.
Fryns JP, Van Buggenhout G (1998). Structural chromosome rearrangements in couples with recurrent fetal wastage. Eur J Obstet Gynecol Reprod Biol. 81:171-176.
Fyfe JC, Madsen M, Hojrup P, Christensen EI, Tanner SM, de la Chapelle A, He Q, Moestrup SK (2004). The functional cobalamin (vitamin B12)-intrinsic factor receptor is a novel complex of cubilin and amnionless. Blood. 103:1573-1579.
Gabut M, Mine M, Marsac C, Brivet M, Tazi J, Soret J (2005). The SR protein SC35 is responsible for aberrant splicing of the E1alpha pyruvate dehydrogenase mRNA in a case of mental retardation with lactic acidosis. Mol Cell Biol. 25:3286-3294.
Gale RE, Fielding AK, Harrison CN, Linch DC (1997). Acquired skewing of X-chromosome inactivation patterns in myeloid cells of the elderly suggests stochastic clonal loss with age. Br J Haematol. 98:512-519.
Ghosh SS, Fahy E, Bodis-Wollner I, Sherman J, Howell N (1996). Longitudinal study of a heteroplasmic 3460 Leber hereditary optic neuropathy family by multiplexed primer-extension analysis and nucleotide sequencing. Am. J. Hum. Genet. 58:325-334.
Girardi G (2005). Heparin treatment in pregnancy loss: Potential therapeutic benefits beyond anticoagulation. J Reprod Immunol. 66:45-51.
Godley LA, Kopp JB, Eckhaus M, Paglino JJ, Owens J, Varmus HE (1996). Wild-type p53 transgenic mice exhibit altered differentiation of the ureteric bud and possess small kidneys. Genes Dev. 10:836-850.
Gräsbeck R, Gordin R, Kanteroi I, Kuhlback B (1960). Selective vitamin B12 malabsorption and proteinuria in young people. Acta Med Scand. 167:289-296.
Grzybowski T (2000). Extremely high levels of human mitochondrial DNA heteroplasmy in single hair roots. Electrophoresis. 21:548-553.
Gu JM, Crawley JT, Ferrell G, Zhang F, Li W, Esmon NL, Esmon CT (2002). Disruption of the endothelial cell protein C receptor gene in mice causes placental thrombosis and early embryonic lethality. J Biol Chem. 277:43335-43343.
Haaf T (2006). Methylation dynamics in the early mammalian embryo: implications of genome reprogramming defects for development. Curr Top Microbiol Immunol. 310:13-22.
Healy AM, Rayburn HB, Rosenberg RD, Weiler H (1995). Absence of the blood-clotting regulator thrombomodulin causes embryonic lethality in mice before development of a functional cardiovascular system. Proc Natl Acad Sci USA. 92:850-854.
Healy AM, Hancock WW, Christie PD, Rayburn HB, Rosenberg RD (1998). Intravascular coagulation activation in a murine model of thrombomodulin deficiency: effects of lesion size, age, and hypoxia on fibrin deposition. Blood. 92:4188-4197.
84
Hiatt BK, Lentz SR (2002). Prothrombotic states that predispose to stroke. Curr Treat Options Neurol. 4(6):417-425.
Hill JA, Choi BC (2000). Maternal immunological aspects of pregnancy success and failure. J Reprod Fertil. 55(Suppl):91–97.
Hogge WA, Prosen TL, Lanasa MC, Huber HA, Reeves MF (2007). Recurrent spontaneous abortion and skewed X-inactivation: is there an association? Am J Obstet Gynecol. 196:384.e1-8.
Hopmeier P, Puehringer H, van Trotsenburg M, Atamaniuk J, Oberkanins C, Dossenbach-Glaninger A (2008). Association of endothelial protein C receptor haplotypes, factor V Leiden and recurrent first trimester pregnancy loss. Clin Biochem. 41:1022-1024.
Hopps CV, Mielnik A, Goldstein M, Palermo GD, Rosenwaks Z, Schlegel PN (2003). Detection of sperm in men with Y chromosome microdeletions of the AZFa, AZFb and AZFc regions. Hum Reprod. 18:1660-1665.
Hoshimoto K, Komine F, Hayashi M, Ohkura T (2002). Plasma soluble Fas changes during early pregnancy and miscarriage. Clin Chim Acta. 323:157-160.
Howell N, Chinnery PF, Ghosh SS, Fahy E, Turnbull DM (2000). Transmission of the human mitochondrial genome. Hum. Reprod., 15(Suppl):235-245.
Iitsuka Y, Bock A, Nguyen DD, Samango-Sprouse CA, Simpson JL, Bischoff FZ (2001). Evidence of skewed X-chromosome inactivation in 47,XXY and 48,XXYY Klinefelter patients. Am J Med Genet. 98:25-31.
Imerslund O (1960). Idiopathic chronic megaloblastic anemia in children. Acta Paediatr. 49(Suppl):1-115.
International Human Genome Sequencing Consortium. Finishing the euchromatic sequence of the human genome (2004). Nature. 431:931-945.
Ireland H, Kunz G, Kyriakoulis K, Stubbs PJ, Lane DA (1997). Thrombomodulin gene mutations associated with myocardial infarction. Circulation. 96:15-18.
Ireland H, Konstantoulas CJ, Cooper JA, Hawe E, Humphries SE, Mather H, Goodall AH, Hogwood J, Juhan-Vague I, Yudkin JS, di Minno G, Margaglione M, Hamsten A, Miller GJ, Bauer KA, Kim YT, Stearns-Kurosawa DJ, Kurosawa S (2005). EPCR Ser219Gly: elevated sEPCR, prothrombin F1+2, risk for coronary heart disease, and increased sEPCR shedding in vitro. Atherosclerosis. 183:283-292.
Isermann B, Hendrickson SB, Hutley K, Wing M, Weiler H (2001). Tissue-restricted expression of thrombomodulin in the placenta rescues thrombomodulin-deficient mice from early lethality and reveals a secondary developmental block. Development. 128:827-838.
Jacks T (1996). Tumor suppressor gene mutations in mice. Annu Rev Genet. 30:603-36.
Jacobi FK, Leo-Kottler B, Mittelviefhaus K, Zrenner E, Meyer J, Pusch CM, Wissinger B (2001). Segregation patterns and heteroplasmy prevalence in Leber's hereditary optic neuropathy. Invest. Ophthalmol. Vis. Sci. 42:1208-1214.
Jenuth JP, Peterson AC, Fu K, Shoubridge EA (1996). Random genetic drift in the female germline explains the rapid segregation of mammalian mitochondrial DNA. Nat. Genet. 14:146-151.
Johnson TM, Hammond EM, Giaccia A, Attardi LD (2005). The p53QS transactivation-deficient mutant shows stress-specific apoptotic activity and induces embryonic lethality. Nat Genet. 37:145-152.
Jones SN, Roe AE, Donehower LA, Bradley A (1995). Rescue of embryonic lethality in Mdm2-deficient mice by absence of p53. Nature. 378:206-208.
Kalantry S, Manning S, Haub O, Tomihara-Newberger C, Lee HG, Fangman J, Disteche CM, Manova K, Lacy E (2001). The amnionless gene, essential for mouse gastrulation, encodes a visceral-endoderm-specific protein with an extracellular cysteine-rich domain. Nat Genet. 27:412-416.
85
Kalousek DK, Pantzar T, Tsai M, Paradice B (1993). Early spontaneous abortion: morphologic and karyotypic findings in 3,912 cases. Birth Defects Orig Artic Ser. 29:53-61.
Katz VL, Kuller JA (1994). Recurrent miscarriage. Am J Perinat. 11:386-397.
Kay C, Jeyendran RS, Coulam CB (2006). p53 tumour suppressor gene polymorphism is associated with recurrent implantation failure. Reprod Biomed Online. 13:492-496.
Kere J (2001) Human population genetics: lessons from Finland. Annu Rev Genomics Hum Genet. 2:103-128.
Kesmodel U, Wisborg K, Olsen SF, Henriksen TB, Secher NJ (2002). Moderate alcohol intake in pregnancy and the risk of spontaneous abortion. Alcohol Alcohol. 37:87-92.
Khrapko K (2008). Two ways to make an mtDNA bottleneck. Nat. Genet. 40:134-135.
Kim JW, Park SY, Kim YM, Kim JM, Han JY, Ryu HM (2004). X-chromosome inactivation patterns in Korean women with idiopathic recurrent spontaneous abortion. J Korean Med Sci. 19:258-262.
Koehler CM, Lindberg GL, Brown DR, Beitz DC, Freeman AE, Mayfield JE, Myers AM (1991). Replacement of bovine mitochondrial DNA by a sequence variant within one generation. Genetics. 129:247-255.
Krausz C, McElreavey K (1999). Y chromosome and male infertility. Front Biosci. 15:E1-8.
Krawczak M, Reiss J, Cooper DN (1992). The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Hum Genet. 90:41-54.
Kujovich JL (2004). Thrombophilia and pregnancy complications. Am J Obstet Gynecol. 191:412-424.
Kuo PL (2002). Maternal trisomy 21 mosaicism and recurrent spontaneous abortion. Fertil Steril 78:432–433.
Kuo PL, Huang SC, Chang LW, Lin CH, Tsai WH, Teng YN (2008). Association of extremely skewed X-chromosome inactivation with Taiwanese women presenting with recurrent pregnancy loss. J Formos Med Assoc. 107:340-343.
Kupferminc M (2003). Thrombophilia and pregnancy. Reprod Biol Endocrinol 1:111.
Kutteh WH, Triplett DA (2006). Thrombophilias and recurrent pregnancy loss. Semin Reprod Med. 24:54-66.
Lacerda LL, Serrano SV, Mathes A, Rey JA, Bello MJ, Casartelli C (2005). An intronic variant in the TP53 gene in a Brazilian woman with breast cancer. Cancer Genet Cytogenet. 160:160-163.
Lanasa MC, Hogge WA, Kubik C, Blancato J, Hoffman EP (1999). Highly skewed X-chromosome inactivation is associated with idiopathic recurrent spontaneous abortion. Am J Hum Genet. 65:252-254.
Lanasa MC, Hogge WA, Kubik CJ, Ness RB, Harger J, Nagel T, Prosen T, Markovic N, Hoffman EP (2001). A novel X chromosome-linked genetic cause of recurrent spontaneous abortion. Am J Obstet Gynecol. 185:563–568.
Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, Devon K, Dewar K, Doyle M, FitzHugh W, Funke R, Gage D, Harris K, Heaford A, Howland J, Kann L, Lehoczky J, LeVine R, McEwan P, McKernan K, Meldrim J, Mesirov JP, Miranda C, Morris W, Naylor J et al. (2001) Initial sequencing and analysis of the human genome. Nature. 409:860-921.
Larsson NG, Tulinius MH, Holme E, Oldfors A, Andersen O, Wahlstrom J, Aasly J (1992). Segregation and manifestations of the mtDNA tRNA(Lys) A-->G(8344) mutation of myoclonus epilepsy and ragged-red fibers (MERRF) syndrome. Am. J. Hum. Genet. 51:1201-1212.
Laskin CA, Bombsrdier C, Hannah ME, Mandel FP, Ritchie JWK, Farewell V, Farine D, Spitzer K, Fielding L, Soloninka C, Yeung M (1997). Prednisone and aspirin in women with autoantibodies and unexplained recurrent fetal loss. N Eng J Med. 337:148-153.
86
Laszik Z, Mitro A, Taylor FB Jr, Ferrell G, Esmon CT (1997). Human protein C receptor is present primarily on endothelium of large blood vessels: implications for the control of the protein C pathway. Circulation. 96:3633-3640.
Lau AW, Brown CJ, Penaherrera M, Langlois S, Kalousek DK, Robinson WP (1997). Skewed X-chromosome inactivation is common in fetuses or newborns associated with confined placental mosaicism. Am J Hum Genet. 61:1353-1361.
Le Flem L, Picard V, Emmerich J, Gandrille S, Fiessinger JN, Aiach M, Alhenc-Gelas M (1999). Mutations in promoter region of thrombomodulin and venous thromboembolic disease. Arterioscler Thromb Vasc Biol. 19:1098-1104.
Lee KY, DeMayo FJ (2004). Animal models of implantation. Reproduction. 128:679-695.
Lehman TA, Haffty BG, Carbone CJ, Bishop LR, Gumbs AA, Krishnan S, Shields PG, Modali R, Turner BC (2000). Elevated frequency and functional activity of a specific germ-line p53 intron mutation in familial breast cancer. Cancer Res. 60:1062-1069.
Levine AJ (1997). p53, the cellular gatekeeper for growth and division. Cell. 88:323-331.
Levine JS, Branch DW, Rauch J (2002). The antiphospholipid syndrome. N Engl J Med. 346(10):752-763.
Levy S, Sutton G, Ng PC, Feuk L, Halpern AL, Walenz BP, Axelrod N, Huang J, Kirkness EF, Denisov G, Lin Y, MacDonald JR, Pang AW, Shago M, Stockwell TB, Tsiamouri A, Bafna V, Bansal V, Kravitz SA, Busam DA, Beeson KY, McIntosh TC, Remington KA, Abril JF, Gill J, Borman J, Rogers YH, Frazier ME, Scherer SW, Strausberg RL, Venter JC (2007). The diploid genome sequence of an individual human. PLoS Biol. 5:e254.
Li TC, Tuckerman EM, Laird SM (2002a). Endometrial factors in recurrent miscarriage. Hum Reprod Update. 8:43-52.
Li TC, Makris M, Tomsu M, Tuckerman EM, Laird SM (2002b). Recurrent miscarriage: aetiology, management and prognosis. Hum Reprod Update. 8:463-481.
Li W, Zheng X, Gu JM, Ferrell GL, Brady M, Esmon NL, Esmon CT (2005). Extraembryonic expression of EPCR is essential for embryonic viability. Blood. 106:2716-2722.
Lim CK, Cho JW, Song IO, Kang IS, Yoon YD, Jun JH (2008) Estimation of chromosomal imbalances in preimplantation embryos from preimplantation genetic diagnosis cycles of reciprocal translocations with or without acrocentric chromosomes. Fertil Steril. [Epub ahead of print]
Lim KS, Naviaux RK, Wong S, Haas RH (2008). Pitfalls in the denaturing high-performance liquid chromatography analysis of mitochondrial DNA mutation. J Mol Diagn. 10:102-108.
Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25: 402-408.
Los FJ, Van Opstal D, van den Berg C (2004) The development of cytogenetically normal, abnormal and mosaic embryos: a theoretical model. Hum Reprod Update. 10:79-94.
Lucifero D, Chaillet JR, Trasler JM (2004). Potential significance of genomic imprinting defects for reproduction and assisted reproductive technology. Hum Reprod Update. 10:3-18.
Luder AS, Tanner SM, de la Chapelle A, Walter JH (2008). Amnionless (AMN) mutations in Imerslund-Gräsbeck syndrome may be associated with disturbed vitamin B(12) transport into the CNS. J Inherit Metab Dis. Jan 7. [Epub ahead of print]
Lyon MF (1961). Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature. 190:372-373.
Lyon MF (1971). Possible mechanisms of X chromosome inactivation. Nat New Biol. 232:229–232.
Malcom S (2005). Genotype to Phenotype. Guide to mutation detection. Published by John Wiley &Sons, Inc.
Martinez-Barbera JP, Beddington RS (2001). Getting your head around Hex and Hesx1: forebrain formation in mouse. Int J Dev Biol. 45:327-336.
87
Maruyama I, Bell CE, Majerus PW (1985). Thrombomodulin is found on endothelium of arteries, veins, capillaries, and lymphatics, and on syncytiotrophoblast of human placenta. J Cell Biol. 101:363-371.
Marzusch K, Ruck P, Horny HP, Dietl J, Kaiserling E (1995). Expression of the p53 tumour suppressor gene in human placenta: an immunohistochemical study. Placenta. 16:101-104.
McFadden DE, Friedman JM (1997) Chromosome abnormalities in human beings. Mutat Res. 396:129-140.
McFarland R, Clark KM, Morris AA, Taylor RW, Macphail S, Lightowlers RN, Turnbull DM (2002). Multiple neonatal deaths due to a homoplasmic mitochondrial DNA mutation. Nat. Genet. 30:145-146.
McKenzie M, Liolitsa D, Hanna MG (2004). Mitochondrial disease: mutations and mechanisms. Neurochem. Res. 29:589-600.
Medina P, Navarro S, Estelles A, Vaya A, Woodhams B, Mira Y, Villa P, Migaud-Fressart M, Ferrando F, Aznar J, Bertina RM, Espana F (2004). Contribution of polymorphisms in the endothelial protein C receptor gene to soluble endothelial protein C receptor and circulating activated protein C levels, and thrombotic risk. Thromb Haemost. 91:905-911.
Medina P, Navarro S, Estelles A, Vaya A, Bertina RM, Espana F (2005). Influence of the 4600A/G and 4678G/C polymorphisms in the endothelial protein C receptor (EPCR) gene on the risk of venous thromboembolism in carriers of factor V Leiden. Thromb Haemost. 94:389-394.
Mills JL, Simpson JL, Driscoll SG, Jovanovic-Petersen L, Van Allen M, Aarons JH, Metzger B, Bieber FR, Knopp RH, Holmes LB (1988). Incidence of spontaneous abortion among normal women and insulin dependent diabetic women whose pregnancies were identified within 21 days of conception. N Eng J Med. 319:1617-1623.
Minks J, Robinson WP, Brown CJ (2008) A skewed view of X chromosome inactivation. J Clin Invest. 118:20-23.
Montes de Oca Luna R, Wagner DS, Lozano G (1995). Rescue of early embryonic lethality in mdm2-deficient mice by deletion of p53. Nature. 378:203-206.
Moomjy M, Colombero LT, Veeck LL, Rosenwaks Z, Palermo GD (1999). Sperm integrity is critical for normal mitotic division and early embryonic development. Mol Hum Reprod. 5:836-844.
Mtiraoui N, Zammiti W, Ghazouani L, Braham NJ, Saidi S, Finan RR, Almawi WY, Mahjoub T (2006) Methylenetetrahydrofolate reductase C677T and A1298C polymorphism and changes in homocysteine concentrations in women with idiopathic recurrent pregnancy losses. Reproduction. 131:395-401.
Munne S, Alikani M, Tomkin G, Grifo J, Cohen J (1995). Embryo morphology, developmental rates, and maternal age correlated with chromosome abnormalities. Fertil Steril. 64:382-391.
Munne S, Escudero T, Sandalinas M, Sable D, Cohen J (2000). Gamete segregation in female carriers of Robertsonian translocations. Cytogenet Cell Genet. 90:303-308.
Mutter GL (1997). Role of imprinting in abnormal human development. Mutat Res. 396:141-147.
Nakai K, Sakamoto H (1994). Construction of a novel database containing aberrant splicing mutations of mammalian genes. Gene. 141:171-177.
Nakazawa F, Koyama T, Shibamiya A, Hirosawa S (2002). Characterization of thrombomodulin gene mutations of the 5'-regulatory region. Atherosclerosis. 164:385-387.
Naumova AK, Plenge RM, Bird LM, Leppert M, Morgan K, Willard HF, Sapienza C (1996). Heritability of X chromosome--inactivation phenotype in a large family. Am J Hum Genet. 58:1111-1119.
Naumova AK, Olien L, Bird LM, Smith M, Verner AE, Leppert M, Morgan K, Sapienza C (1998). Genetic mapping of X-linked loci involved in skewing of X chromosome inactivation in the human. Eur J Hum Genet. 6:552-562.
88
Nicol CJ, Harrison ML, Laposa RR, Gimelshtein IL, Wells PG (1995). A teratologic suppressor role for p53 in benzo[a]pyrene-treated transgenic p53-deficient mice. Nat Genet. 10:181-187.
Nobrega MA, Pennacchio LA (2004). Comparative genomic analysis as a tool for biological discovery. J Physiol. 554:31-39.
Nolan PM, Hugill A, Cox RD (2002). ENU mutagenesis in the mouse: application to human genetic disease.Brief Funct Genomic Proteomic. 1:278-289.
Noordam MJ, van der Veen F, Repping S (2006). Techniques and reasons to remain interested in the Y chromosome. Fertil Steril. 86:1801-1802.
Noordam MJ, Repping S (2006). The human Y chromosome: a masculine chromosome. Curr Opin Genet Dev. 16:225-232.
Norimura T, Nomoto S, Katsuki M, Gondo Y, Kondo S (1996). p53-dependent apoptosis suppresses radiation-induced teratogenesis. Nat Med. 2:577-580.
Norio R (2003a). Finnish Disease Heritage I: characteristics, causes, background. Hum Genet. 112:441-456.
Norio R (2003b). Finnish Disease Heritage II: population prehistory and genetic roots of Finns. Hum Genet. 112:457-469.
Norwitz ER, Schust DJ, Fisher SJ (2001). Implantation and the survival of early pregnancy. N Engl J Med. 345:1400-1408.
Ntrivalas EI, Kwak-Kim JY, Gilman-Sachs A, Chung-Bang H, Ng SC, Beaman KD, Mantouvalos HP, Beer AE (2001). Status of peripheral blood natural killer cells in women with recurrent spontaneous abortions and infertility of unknown aetiology. Hum Reprod. 16(5):855-861.
Nybo Andersen AM, Wohlfahrt J, Christens P, Olsen J, Melbye M (2000). Maternal age and fetal loss: population based register linkage study. BMJ. 320:1708-1712.
Ober C, Aldrich CL, Chervoneva I, Billstrand C, Rahimov F, Gray HL, Hyslop T (2003). Variation in the HLA-G promoter region influences miscarriage rates. Am J Hum Genet. 72:1425-1435.
Ogasawara M, Aoki K, Okada S, Suzumori K (2000). Embryonic karyotype of abortuses in relation to the number of previous miscarriages. Fertil Steril. 73:300-304.
Okon MA, Laird SM, Tuckerman EM, Li TC (1998). Serum androgen levels in women who have recurrent miscarriages and their correlation with markers of endometrial function. Fertil Steril. 69:682-690.
Olivo PD, Van de Walle MJ, Laipis PJ, Hauswirth WW (1983). Nucleotide sequence evidence for rapid genotypic shifts in the bovine mitochondrial DNA D-loop. Nature. 306:400-402.
Otani T, Roche M, Mizuike M, Colls P, Escudero T, Munné S (2006) Preimplantation genetic diagnosis significantly improves the pregnancy outcome of translocation carriers with a history of recurrent miscarriage and unsuccessful pregnancies. Reprod Biomed Online. 13:869-874.
Pandey MK, Thakur S, Agrawal S (2004). Lymphocyte immunotherapy and its probable mechanism in the maintance of precnancy in women with recurrent spontaneous abortion. Arch Gynecol Obstet. 269:161-172.
Pandey MK, Rani R, Agrawal S (2005). An update in recurrent spontaneous abortion. Arch Gynecol Obstet. 272:95-108.
Parazzini F, Chatenoud L, Di Cintio E, Mezzopane R, Surace M, Zanconato G, Fedele L, Benzi G (1998). Coffee consumption and risk of hospitalized miscarriage before 12 weeks of gestation. Hum Reprod. 13:2286-2291.
Pegoraro E, Whitaker J, Mowery-Rushton P, Surti U, Lanasa M, Hoffman EP (1997). Familial skewed X inactivation: a molecular trait associated with high spontaneous-abortion rate maps to Xq28. Am J Hum Genet. 61:160-170.
Peltonen L, Jalanko A, Varilo T (1999) Molecular genetics of the Finnish disease heritage. Hum Mol Genet. 8:1913-1923.
89
Peltonen L, McKusick VA (2001). Genomics and medicine. Dissecting human disease in the postgenomic era. Science. 291:1224-1229.
Percec I, Plenge RM, Nadeau JH, Bartolomei MS, Willard HF (2002). Autosomal dominant mutations affecting X inactivation choice in the mouse. Science. 296:1136-1139.
Perea-Gomez A, Vella FD, Shawlot W, Oulad-Abdelghani M, Chazaud C, Meno C, Pfister V, Chen L, Robertson E, Hamada H, Behringer RR, Ang SL (2002). Nodal antagonists in the anterior visceral endoderm prevent the formation of multiple primitive streaks. Dev Cell. 3:745-756.
Peters LL, Robledo RF, Bult CJ, Churchill GA, Paigen BJ, Svenson KL (2007). The mouse as a model for human biology: a resource guide for complex trait analysis. Nat Rev Genet. 8:58-69.
Pietrowski D, Bettendorf H, Riener EK, Keck C, Hefler LA, Huber JC, Tempfer C (2005). Recurrent pregnancy failure is associated with a polymorphism in the p53 tumour suppressor gene. Hum Reprod. 20:848-851.
Piette J, Neel H, Marechal V (1997). Mdm2: keeping p53 under control. Oncogene. 15:1001-1010.
Pim D, Banks L (2004). p53 polymorphic variants at codon 72 exert different effects on cell cycle progression. Int J Cancer. 108:196-199.
Plouffe L, White EW, Tho SP, Sweet CS, Layman LC, Whitman GF, McDonough PG (1992). Etiologic factors of recurrent abortion and subsequent reproductive performance of couples: Have we made any progress in the past 10 years? Am J Obstet Gynecol. 67:313-321.
Poland BJ, Miller JR, Jones DC, Trimble BK (1977). Reproductive counseling in patients who have had a spontaneous abortion. Am J Obstet Gynecol. 127:685-691.
Potdar N, Konje JC (2005).The endocrinological basis of recurrent miscarriages. Curr Opin Obstet Gynecol. 17:424-428.
Preston FE, Rosendaal FR, Walker ID, Breit E, Berntorp E, Conard J, Fontcuberta J, Makris M, Mariani G, Noteboom W, Pabinger I, Legnani C, Scharrer I, van der Meer FJ (1996). Increased fetal loss in women with heritable thrombophilia. Lancet. 348:913-916.
Prigoshin N, Tambutti M, Larriba J, Gogorza S, Testa R (2004). Cytokine gene polymorphisms in recurrent pregnancy loss of unknown cause. Am J Reprod Immunol. 52:36-41.
Puck JM, Willard HF (1998). X inactivation in females with X-linked disease. N Engl J Med. 338:325-328.
Quenby S, Vince G, Farquharson R, Aplin J (2002). Recurrent miscarriage: a defect in nature's quality control? Hum Reprod. 17:1959-1963.
Rai R, Regan L (2006). Recurrent miscarriage. Lancet. 368:601-611.
Rajasimha HK, Chinnery PF, Samuels DC (2008). Selection against pathogenic mtDNA mutations in a stem cell population leads to the loss of the 3243A-->G mutation in blood. Am J Hum Genet. 82:333-343.
Rand JH, Wu X-X, Andree HA, Lockwood CJ, Guller S, Scher J, Harpel PC (1997). Pregnancy loss in the antiphospholipid-antibody syndrome- a possible thrombogenic mechanism. N Eng J Med. 337(3):154-160.
Rebbeck TR, Spitz M, Wu X (2004). Assessing the function of genetic variants in candidate gene association studies. Nat Rev Genet. 5:589-597.
Regan L (1991). Recurrent miscarriage. BMJ. 302:543-544.
Regan L, Braude PR, Trembath PL (1989). Influence of past reproductive performance on risk of spontaneous abortion. BMJ. 299:541-545.
Reik W, Dean W (2001). DNA methylation and mammalian epigenetics. Electrophoresis. 22:2838-2843.
Reik W, Dean W, Walter J (2001). Epigenetic reprogramming in mammalian development. Science. 293:1089-1093.
90
Reinhard G, Noll A, Schlebusch H, Mallmann P, Ruecker AV. Shifts in the TH1/TH2 balance during human pregnancy correlate with apoptotic changes (1998). Biochem Biophys Res Commun. 245(3):933-938.
Rey E, Kahn SR, David M, Shrier I (2003). Thrombophilic disorders and fetal loss: a meta-analysis. Lancet 361:901–908.
Reznikoff-Etiévan MF, Cayol V, Carbonne B, Robert A, Coulet F, Milliez J (2001) Factor V Leiden and G20210A prothrombin mutations are risk factors for very early recurrent miscarriage.BJOG. 108:1251-1254.
Risch HA, Weiss NS, Clarke EA, Miller AB (1988). Risk factors for spontaneous abortion and its recurrence. Am J Epidemiol. 128:420-430.
Roberts CJ, Lowe CR (1975). Where have all the conceptions gone? Lancet. i: 498-501.
Roberts CP, Murphy AA (2000). Endocrinopathies associated with recurrent pregnancy loss. Semin Reprod Med. 18:357–362.
Rogel A, Popliker M, Webb CG, Oren M (1985). p53 cellular tumor antigen: analysis of mRNA levels in normal adult tissues, embryos, and tumors. Mol Cell Biol. 5:2851-2855.
Rossant J, Cross JC (2001). Placental development: lessons from mouse mutants. Nat Rev Genet. 2:538-548.
Rubio C, Simon C, Blanco J, Vidal F, Minguez Y, Egozcue J, Crespo J, Remohi J, Pellicer A (1999). Implications of sperm chromosome abnormalities in recurrent miscarriage. J Assist Reprod Genet. 16:253-258.
Ruzzi L, Ciarafoni I, Silvestri L, Semeraro ML, Abeni D (2005). Association of PLA2 polymorphism of the ITGB3 gene with early fetal loss. Fertil Steril. 83:511-512.
Sadler TW, Langman J, Leland J (2006). Langman´s Medical Embryology. 10th edition, published
by Lippincott Williams and Wilkins.
Sah VP, Attardi LD, Mulligan GJ, Williams BO, Bronson RT, Jacks T (1995). A subset of p53-deficient embryos exhibit exencephaly. Nat Genet. 10:175-180.
Santos F, Dean W (2004). Epigenetic reprogramming during early development in mammals. Reproduction. 127:643-651.
Sariola H, Frilander M, Heino T, Jernvall J, Partanen J, Sainio K, Salminen M, Thesleff I (2003). 1
st edition, published by Gummerus Kirjapaino Oy.
Schmid P, Lorenz A, Hameister H, Montenarh M (1991). Expression of p53 during mouse embryogenesis. Development. 113:857-865.
Sharp A, Robinson D, Jacobs P (2000). Age- and tissue-specific variation of X chromosome inactivation ratios in normal women. Hum Genet. 107:343–349.
Shoubridge EA (2000). Mitochondrial DNA segregation in the developing embryo. Hum. Reprod. 15(Suppl):229-234.
Simerly C, Wu GJ, Zoran S, Ord T, Rawlins R, Jones J, Navara C, Gerrity M, Rinehart J, Binor Z (1995). The paternal inheritance of the centrosome, the cell's microtubule-organizing center, in humans, and the implications for infertility. Nat Med. 1:47-52.
Simmonds RE, Lane DA (1999). Structural and functional implications of the intron/exon organization of the human endothelial cell protein C/activated protein C receptor (EPCR) gene: comparison with the structure of CD1/major histocompatibility complex alpha1 and alpha2 domains. Blood. 94:632-641.
Simoni M, Bakker E, Krausz C (2004). EAA/EMQN best practice guidelines for molecular diagnosis of y-chromosomal microdeletions. Int J Androl. 27:240-249.
Snijders RJ, Sundberg K, Holzgreve W, Henry G, Nicolaides KH (1999). Maternal age- and gestation-specific risk for trisomy 21. Ultrasound Obstet Gynecol. 13:167-170.
Souza SS, Ferriani RA, Pontes AG, Zago MA, Franco RF (1999) Factor V leiden and factor II G20210A mutations in patients with recurrent abortion. Hum Reprod. 14:2448-1250.
91
Stearns-Kurosawa DJ, Kurosawa S, Mollica JS, Ferrell GL, Esmon CT (1996). The endothelial cell protein C receptor augments protein C activation by the thrombin-thrombomodulin complex. Proc Natl Acad Sci U S A. 93:10212-10216.
Stephenson MD, Dreher K, Houlihan E, Wu W (1998). Prevention of unexplained recurrent spontaneous abortion using intravenous immunoglobulin: a prospective, randomized, double-blinded, placebo-controlled trial. Am J Reprod Immunol. 39:82-88.
Stephenson MD, Awartani KA, Robinson WP (2002). Cytogenetic analysis of miscarriages from couples with recurrent miscarriage: a case-control study. Hum Reprod. 17:446-451.
Stephenson M, Kutteh W (2007). Evaluation and management of recurrent early pregnancy loss. Clin Obstet Gynecol. 50:132-145.
Stern JJ, Dorfman AD, Gutierrez-Najar AJ, Cerrillo M, Coulam CB (1996). Frequency of abnormal karyotypes among abortuses from women with and without a history of recurrent spontaneous abortion. Fertil Steril. 65:250-253.
Stewart JB, Freyer C, Elson JL, Wredenberg A, Cansu Z, Trifunovic A, Larsson NG (2008). Strong purifying selection in transmission of mammalian mitochondrial DNA. PLoS Biol. 6:e10.
Stirrat GM (1990). Recurrent miscarriage. Lancet. 336:673-675.
Stogmann E, Lichtner P, Baumgartner C, Schmied M, Hotzy C, Asmus F, Leutmezer F, Bonelli S, Assem-Hilger E, Vass K, Hatala K, Strom TM, Meitinger T, Zimprich F, Zimprich A (2006). Mutations in the CLCN2 gene are a rare cause of idiopathic generalized epilepsy syndromes. Neurogenetics. 7:265-268.
Strachan T, Read AP (2004). Human Molecular Genetics. Third Edition, published by Garland Science.
Stray-Pedersen B, Stray-Pedersen S (1984). Etiologic factors and subsequent reproductive performance in 195 couples with a prior history of habitual abortion. Am J Obstet Gynecol. 148:140-146.
Strope S, Rivi R, Metzger T, Manova K, Lacy E (2004). Mouse amnionless, which is required for primitive streak assembly, mediates cell-surface localization and endocytic function of cubilin on visceral endoderm and kidney proximal tubules. Development. 131:4787-4795.
Suh Y, Vijg J (2005). SNP discovery in associating genetic variation with human disease phenotypes. Mutat Res. 573:41-53.
Sullivan AE, Lewis T, Stephenson M, Odem R, Schreiber J, Ober C, Branch DW (2003). Pregnancy outcome in recurrent miscarriage patients with skewed X chromosome inactivation. Obstet Gynecol. 101:1236-1242.
Sullivan AE, Silver RM, LaCoursiere DY, Porter TF, Branch DW (2004). Recurrent fetal aneuploidy and recurrent miscarriage. Obstet Gynecol. 104:784-788.
Suomalainen A, Syvänen AC (2000) Quantitative analysis of human DNA sequences by PCR and solid-phase minisequencing. Mol Biotechnol 15:123-131.
Suryanarayana V, Rao L, Kanakavalli M, Padmalatha V, Deenadayal M, Singh L (2006). Recurrent early pregnancy loss and endothelial nitric oxide synthase gene polymorphisms. Arch Gynecol Obstet. 274:119-124.
Swales AK, Spears N (2005). Genomic imprinting and reproduction. Reproduction. 130:389-399.
Takakuwa K, Adachi H, Hataya I, Ishii K, Tamura M, Tanaka K (2003). Molecular genetic studies of HLA-DRB1 alleles in patients with unexplained recurrent abortion in the Japanese population. Hum Reprod. 18:728-733.
Tal J, Kaplan M, Sharf M, Barnea ER (1991). Stress-related hormones affect human chorionic gonadothrophin secretion from the early human placenta in vitro. Hum Reprod. 6:766-769.
Tanner SM, Aminoff M, Wright FA, Liyanarachchi S, Kuronen M, Saarinen A, Massika O, Mandel H, Broch H, de la Chapelle A (2003). Amnionless, essential for mouse gastrulation, is mutated in recessive hereditary megaloblastic anemia. Nat Genet. 33:426-429.
92
Tanner SM, Li Z, Bisson R, Acar C, Oner C, Oner R, Cetin M, Abdelaal MA, Ismail EA, Lissens W, Krahe R, Broch H, Grasbeck R, de la Chapelle A (2004). Genetically heterogeneous selective intestinal malabsorption of vitamin B12: founder effects, consanguinity, and high clinical awareness explain aggregations in Scandinavia and the Middle East. Hum Mut. 23:327-233.
Tay SK, Shanske S, Kaplan P, DiMauro S (2004). Association of mutations in SCO2, a cytochrome c oxidase assembly gene, with early fetal lethality. Arch Neurol. 61:950-952.
Taylor RW, Turnbull DM (2005). Mitochondrial DNA mutations in human disease. Nat. Rev. Genet. 6:389-402.
Teebi AS, Teebi SA (2005). Genetic diversity among the Arabs. Community Genet. 8:21-26.
Thomas M, Kalita A, Labrecque S, Pim D, Banks L, Matlashewski G (1999). Two polymorphic variants of wild-type p53 differ biochemically and biologically. Mol Cell Biol. 19:1092-1100.
Thorburn DR, Dahl HH (2001). Mitochondrial disorders: genetics, counseling, prenatal diagnosis and reproductive options. Am. J. Med. Genet. 106:102-114.
Tomihara-Newberger C, Haub O, Lee HG, Soares V, Manova K, Lacy E (1998). The amn gene product is required in extraembryonic tissues for the generation of middle primitive streak derivatives. Dev Biol. 204:34-54.
Tuckey RC (2005). Progesterone synthesis by the human placenta. Placenta. 26:273-281.
Tulppala M, Palosuo T, Ramsay T, Miettinen, Salonen R, Ylikorkala O (1993). A prospective study of 63 couples with history of recurrent spontaneous abortion: contributing factors and outcome of subsequent pregnancies. Hum Reprod. 8:764-770.
Uehara S, Hashiyada M, Sato K, Sato Y, Fujimori K, Okamura K (2001). Preferential X-chromosome inactivation in women with idiopathic recurrent pregnancy loss. Fertil Steril. 76:908-914.
Van Blerkom J (1996). Sperm centrosome dysfunction: a possible new class of male factor infertility in the human. Mol Hum Reprod. 2:349-354.
Van Blerkom J (2004). Mitochondria in human oogenesis and preimplantation embryogenesis: engines of metabolism, ionic regulation and developmental competence. Reproduction. 128:269-280.
Van Blerkom J, Sinclair J, Davis P (1998). Mitochondrial transfer between oocytes: potential applications of mitochondrial donation and the issue of heteroplasmy. Hum Reprod. 13:2857-2868.
Van Blerkom J, Davis P, Alexander S (2000). Differential mitochondrial distribution in human pronuclear embryos leads to disproportionate inheritance between blastomeres: relationship to microtubular organization, ATP content and competence. Hum Reprod. 15:2621-2633.
Van de Wouwer M, Collen D, Conway EM (2004). Thrombomodulin-protein C-EPCR system: integrated to regulate coagulation and inflammation. Arterioscler Thromb Vasc Biol. 24:1374-1383.
Van der Velden PA, Krommenhoek-Van Es T, Allaart CF, Bertina RM, Reitsma PH (1991). A frequent thrombomodulin amino acid dimorphism is not associated with thrombophilia. Thromb Haemost. 65:511-513.
Varley JM, Chapman P, McGown G, Thorncroft M, White GR, Greaves MJ, Scott D, Spreadborough A, Tricker KJ, Birch JM, Evans DG, Reddel R, Camplejohn RS, Burn J, Boyle JM (1998). Genetic and functional studies of a germline TP53 splicing mutation in a Li-Fraumeni-like family. Oncogene. 16:3291-3298.
Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, Smith HO, Yandell M, Evans CA, Holt RA, Gocayne JD, Amanatides P, Ballew RM, Huson DH, Wortman JR, Zhang Q, Kodira CD, Zheng XH, Chen L, Skupski M, Subramanian G, Thomas PD, Zhang J et al. (2001) The sequence of the human genome. Science. 291:1304-1351.
Vitiello D, Patrizio P (2007). Implantation and early embryonic development: implications for pregnancy. Semin Perinatol. 31:204-207.
93
Voglino G, Castello S, Silengo L, Stefanuto G, Friard O, Ferrara G, Fessia L (1997). An intronic deletion in TP53 gene causes exon 6 skipping in breast cancer. Eur J Cancer. 33:1479-1483.
Vogt PH, Edelmann A, Kirsch S, Henegariu O, Hirschmann P, Kiesewetter F, Kohn FM, Schill WB, Farah S, Ramos C, Hartmann M, Hartschuh W, Meschede D, Behre HM, Castel A, Nieschlag E, Weidner W, Grone HJ, Jung A, Engel W, Haidl G (1996). Human Y chromosome azoospermia factors (AZF) mapped to different subregions in Yq11. Hum Mol Genet. 5:933-943.
Von Linsingen R, Bompeixe EP, Bicalho G (2005). A case-control study in IL6 and TGFB1 gene polymorphisms and recurrent spontaneous abortion in southern Brazilian patients. Am J Reprod Immunol. 53:94-99.
Wan J, Carr JR, Baloh RW, Jen JC (2005). Nonconsensus intronic mutations cause episodic ataxia. Ann Neurol. 57:131-135.
Wang GJ, Yang P, Xie HG (2006). Gene variants in noncoding regions and their possible consequences. Pharmacogenomics. 7:203-209.
Wang H, Dey SK (2006). Roadmap to embryo implantation: clues from mouse models. Nat Rev Genet. 7:185-199.
Wang JX, Davies MJ, Norman RJ (2002). Obesity increases the risk of spontaneous abortion during infertility treatment. Obes Res. 10:551-554.
Wang X, Bornslaeger EA, Haub O, Tomihara-Newberger C, Lonberg N, Dinulos MB, Disteche CM, Copeland N, Gilbert DJ, Jenkins NA, Lacy E, (1996). A candidate gene for the amnionless gastrulation stage mouse mutation encodes a TRAF-related protein. Dev Biol. 177:274-290.
Wang X, Lin Q, Ma Z, Hong Y, Zhao A, Di W, Lu P (2005). Association of the A/G polymorphism at position 49 in exon 1 of CTLA-4 with the susceptibility to unexplained recurrent spontaneous abortion in the Chinese population. Am J Reprod Immunol. 53:100-105.
Wang XB, Zhao X, Giscombe R, Lefvert AK (2002). A CTLA-4 gene polymorphism at position -318 in the promoter region affects the expression of protein. Genes Immun. 3:233-234.
Wang-Gohrke S, Weikel W, Risch H, Vesprini D, Abrahamson J, Lerman C, Godwin A, Moslehi R, Olipade O, Brunet JS, Stickeler E, Kieback DG, Kreienberg R, Weber B, Narod SA, Runnebaum IB (1999). Intron variants of the p53 gene are associated with increased risk for ovarian cancer but not in carriers of BRCA1 or BRCA2 germline mutations. Br J Cancer. 81:179-183.
Warren JE, Silver RM (2008). Genetics of pregnancy loss. Clin Obstet Gynecol. 51:84-95.
Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P, Agarwala R, Ainscough R, Alexandersson M, An P, Antonarakis SE, Attwood J, Baertsch R, Bailey J, Barlow K, Beck S, Berry E, Birren B, Bloom T, Bork P, Botcherby M, Bray N, Brent MR, Brown DG et al. (2002). Initial sequencing and comparative analysis of the mouse genome. Nature. 420:520-562.
Wegmann TG, Lin H, Guilbert L, Mosmann TR (1993). Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a TH2 phenomenon? Immunol Today. 14(7):353-356.
Weiler-Guettler H, Aird WC, Rayburn H, Husain M, Rosenberg RD (1996). Developmentally regulated gene expression of thrombomodulin in postimplantation mouse embryos. Development. 122:2271-2281.
Weinstein LB (2007). Selected genetic disorders affecting Ashkenazi Jewish families. Fam Community Health. 30:50-62.
Wilcox AJ, Weinberg CR, O'Connor JF, Baird DD, Schlatterer JP, Canfield RE, Armstrong EG, Nisula BC (1988). Incidence of early loss of pregnancy. N Engl J Med. 319:189-194.
Wong LJ (2007). Pathogenic mitochondrial DNA mutations in protein-coding genes. Muscle Nerve. 36:279-293.
94
Xiao W, Oefner PJ (2001). Denaturing high-performance liquid chromatography. Hum Mutat. 17:439-474.
Yamada H, Kato EH, Kobashi G, Ebina Y, Shimada S, Morikawa M, Sakuragi N, Fujimoto S (2001). High NK cell activity in early pregnancy correlates with subsequent abortion with normal chromosomes in women with recurrent spontaneous abortion. Am J Reprod Immunol. 46(2):132-136.
Yang Y, Ma M, Li L, Zhang W, Xiao C, Li S, Ma Y, Tao D, Liu Y, Lin L, Zhang S (2008). Evidence for the association of Y-chromosome haplogroups with susceptibility to spermatogenic failure in a Chinese Han population. Med Genet. 45:210-215.
Zhu M, Zhao S (2007) Candidate gene identification approach: progress and challenges. Int J Biol Sci. 3:420-427.